From c149151de5066d33d249f7614f482059a0c4a9ff Mon Sep 17 00:00:00 2001 From: Research Assistant Date: Sat, 13 Jun 2026 19:27:38 +0800 Subject: [PATCH] test: add real-paper trace vs expectations regression harness with fixtures --- .../ocr_real_papers/CAQNW9Q2/block_trace.csv | 156 ++++++++ .../CAQNW9Q2/expectations.json | 151 +++++++- .../ocr_real_papers/DWQQK2YB/block_trace.csv | 288 +++++++++++++++ .../DWQQK2YB/expectations.json | 243 +++++++++++- .../ocr_real_papers/regenerate_traces.py | 130 +++++++ tests/test_ocr_trace_vs_expectations.py | 346 ++++++++++++++++++ 6 files changed, 1290 insertions(+), 24 deletions(-) create mode 100644 tests/fixtures/ocr_real_papers/CAQNW9Q2/block_trace.csv create mode 100644 tests/fixtures/ocr_real_papers/DWQQK2YB/block_trace.csv create mode 100644 tests/fixtures/ocr_real_papers/regenerate_traces.py create mode 100644 tests/test_ocr_trace_vs_expectations.py diff --git a/tests/fixtures/ocr_real_papers/CAQNW9Q2/block_trace.csv b/tests/fixtures/ocr_real_papers/CAQNW9Q2/block_trace.csv new file mode 100644 index 00000000..d2d5b078 --- /dev/null +++ b/tests/fixtures/ocr_real_papers/CAQNW9Q2/block_trace.csv @@ -0,0 +1,156 @@ +page,block_id,raw_label,content_preview,bbox,role,role_confidence,evidence,seed_role,seed_confidence,zone,style_family,marker_type,render_default,index_default +1,0,number,268,"[113, 68, 146, 87]",noise,0.9,page number label,noise,0.9,frontmatter_main_zone,support_like,short_fragment,False,False +1,1,header,Annals of the Rheumatic Diseases 1994; 53: 268–275,"[740, 66, 1108, 88]",noise,0.9,header label,noise,0.9,frontmatter_main_zone,support_like,none,False,False +1,2,paragraph_title,REVIEW,"[112, 135, 218, 162]",unknown_structural,0.7,unnumbered paragraph_title in title zone on page 1: REVIEW,paper_title,0.7,frontmatter_main_zone,heading_like,short_fragment,False,True +1,3,doc_title,Quantitative radiography of osteoarthritis,"[315, 202, 963, 242]",unknown_structural,0.6,page-1 frontmatter title guard: Quantitative radiography of osteoarthritis,paper_title,0.6,frontmatter_main_zone,support_like,none,False,True +1,4,text,J C Buckland-Wright,"[312, 289, 504, 315]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,frontmatter_main_zone,support_like,short_fragment,False,True +1,5,footer,"Division of Anatomy and Cell Biology, United Medical and Dental Schools of Guy's and St Thomas's Hos","[112, 1346, 298, 1493]",noise,0.9,footer label,noise,0.9,frontmatter_main_zone,support_like,none,False,False +1,6,text,Radiography is important in the diagnosis of osteoarthritis (OA) as the features described in the pa,"[313, 357, 706, 916]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,body_like,none,True,True +1,7,text,"Scoring systems, although an essential and widely used method for assessing disease progression, suf","[313, 916, 706, 1383]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,body_like,none,True,True +1,8,footnote,"Correspondence to: Dr J C Buckland-Wright, Division of Anatomy and Cell Biology, UMDS, Guy's Hospita","[114, 1496, 275, 1610]",frontmatter_noise,0.8,"page-1 zone journal_furniture_zone: Correspondence to: +Dr J C Buckland-Wright, +Division of Anato",frontmatter_noise,0.8,frontmatter_main_zone,support_like,none,False,False +1,9,text,Quantitative assessments of the structural changes in peripheral joints with OA are based on measure,"[314, 1382, 706, 1610]",frontmatter_noise,0.8,page-1 zone journal_furniture_zone: Quantitative assessments of the structural changes in periph,frontmatter_noise,0.8,frontmatter_main_zone,body_like,none,False,False +1,10,text,,"[718, 358, 1112, 662]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,frontmatter_main_zone,support_like,empty,False,True +1,11,paragraph_title,Standard radiography,"[720, 700, 932, 722]",unknown_structural,0.5,unnumbered paragraph_title on page 1 outside title zone: Standard radiography,section_heading,0.5,frontmatter_main_zone,heading_like,none,False,True +1,12,text,The radiographic image is a shadow of the differential absorption of x rays by the tissues of the jo,"[718, 722, 1112, 851]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,True,True +1,13,text,"Advantages Standard radiography is simple, cheap, easily accessible and well understood. The radiogr","[719, 850, 1112, 1022]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,True,True +1,14,text,Limitations The relatively large size of the x ray source of standard x ray tubes (usually 1 mm and ,"[719, 1022, 1112, 1298]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,True,True +1,15,text,Apart from those investigators who have developed special stereotaxic devices for examining leg alig,"[719, 1298, 1114, 1609]",frontmatter_noise,0.8,page-1 zone journal_furniture_zone: Apart from those investigators who have developed special st,frontmatter_noise,0.8,frontmatter_main_zone,support_like,none,False,False +2,0,header,Quantitative radiography of OA,"[88, 62, 310, 84]",noise,0.9,header label,noise,0.9,frontmatter_side_zone,support_like,none,False,False +2,1,number,269,"[1057, 61, 1089, 79]",noise,0.9,page number label,noise,0.9,frontmatter_side_zone,support_like,short_fragment,False,False +2,2,image,,"[294, 111, 1087, 755]",media_asset,0.85,media label: image,media_asset,0.85,body_zone,body_like,empty,True,True +2,3,figure_title,Figure 1 Standard radiographic appearance of osteoarthritic knee joints showing different degrees of,"[287, 770, 1088, 842]",figure_caption_candidate,0.92,figure_title label: Figure 1 Standard radiographic appearance of osteoarthritic ,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +2,4,text,reliable assessment of joint space loss. The absence of any standards in the radio-anatomical positi,"[288, 867, 685, 1613]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +2,5,text,"but more often than not the plane of measurement is never defined. Further, no account is taken of t","[694, 866, 1092, 1364]",frontmatter_noise,0.6,default body_paragraph for text label; frontmatter_side_zone excluded from body flow,frontmatter_noise,0.6,frontmatter_side_zone,support_like,none,False,False +2,6,paragraph_title,Quantitative standard radiography,"[697, 1402, 1027, 1423]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Quantitative standard radiography",subsection_heading,0.6,body_zone,heading_like,none,True,True +2,7,text,Quantitative standard radiography has been applied primarily to joint space width (JSW) measurements,"[696, 1421, 1092, 1613]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +3,0,number,270,"[116, 66, 149, 86]",noise,0.9,page number label,noise,0.9,body_zone,body_like,short_fragment,False,False +3,1,header,Buckland-Wright,"[992, 68, 1112, 89]",noise,0.9,header label,noise,0.9,body_zone,body_like,short_fragment,False,False +3,2,text,standardisation of the position of the hip and knee and reproducible repositioning of the joints on ,"[315, 124, 708, 381]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +3,3,text,"In the former, Martel's group at Michigan University, undertook to define the precision of hyaline c","[314, 380, 708, 1237]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +3,4,text,The development of microcomputers and image analysis technique has provided the means of obtaining a,"[313, 1237, 707, 1613]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +3,5,text,,"[719, 126, 1116, 728]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,body_zone,body_like,empty,False,True +3,6,paragraph_title,Microfocal radiography,"[722, 766, 948, 789]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Microfocal radiography",subsection_heading,0.6,body_zone,heading_like,none,True,True +3,7,text,Microfocal x ray units are characterised by an extremely small x ray source (<15 µm in diameter) whi,"[720, 788, 1113, 1025]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +3,8,text,Advantages of microfocal radiography are those characteristic of an extremely small x ray source. La,"[719, 1027, 1114, 1611]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,0,header,Quantitative radiography of OA,"[91, 62, 312, 85]",noise,0.9,header label,noise,0.9,body_zone,body_like,none,False,False +4,1,number,271,"[1059, 58, 1090, 77]",noise,0.9,page number label,noise,0.9,body_zone,body_like,short_fragment,False,False +4,2,image,,"[95, 114, 677, 425]",media_asset,0.85,media label: image,media_asset,0.85,body_zone,unknown_like,empty,True,True +4,3,image,,"[95, 433, 681, 736]",media_asset,0.85,media label: image,media_asset,0.85,body_zone,body_like,empty,True,True +4,4,figure_title,Figure 2 Part of the macroradiographs of osteoarthritic knee joints with medial compartment involvem,"[92, 753, 681, 840]",figure_caption_candidate,0.92,figure_title label: Figure 2 Part of the macroradiographs of osteoarthritic knee,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +4,5,text,"resolution, makes it possible to detect structural detail virtually at the histological level $ ^{34","[295, 865, 687, 973]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,6,text,Limitations are also a function of the small x ray source size. The smallness of the source limits t,"[295, 972, 688, 1122]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,7,text,Care and accuracy are needed in positioning the patient in relation to the source. This requires spe,"[295, 1121, 689, 1380]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,8,paragraph_title,Standardisation of macroradiographic procedure,"[296, 1417, 656, 1459]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Standardisation of macroradiographic procedure",subsection_heading,0.6,body_zone,heading_like,none,True,True +4,9,text,Stereotaxic devices are used to position each patient accurately and reproducibly. The centre of the,"[295, 1459, 689, 1607]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,10,text,,"[698, 113, 1095, 588]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,body_zone,body_like,empty,False,True +4,11,text,The anatomical sites within the macroradiograph used in defining the boundaries for the measurement ,"[700, 586, 1094, 757]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,12,text,Femur: the distal convex margin of the condyle (fig 2).,"[703, 758, 1093, 800]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,13,text,"Tibia, medial compartment: a line extending from near the tibial spine to the medial or outer margin","[701, 800, 1096, 1249]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,14,text,A detailed description of the method of measuring radiographic features and the accuracy in recordin,"[701, 1248, 1098, 1607]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,0,number,272,"[114, 59, 147, 78]",noise,0.9,page number label,noise,0.9,body_zone,body_like,short_fragment,False,False +5,1,header,Buckland-Wright,"[989, 61, 1109, 82]",noise,0.9,header label,noise,0.9,body_zone,body_like,short_fragment,False,False +5,2,paragraph_title,Quantitative microfocal radiography,"[314, 116, 658, 139]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Quantitative microfocal radiography",subsection_heading,0.6,body_zone,heading_like,none,True,True +5,3,text,Measurement of the radiographic features of OA of the hand and their change over an 18 month study p,"[313, 139, 706, 548]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,4,paragraph_title,Joint space width,"[313, 585, 480, 607]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Joint space width",subsection_heading,0.6,body_zone,heading_like,short_fragment,True,True +5,5,text,Joint space narrowing is considered the most important radiological feature of OA but its accuracy i,"[311, 608, 706, 906]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,6,text,"In patients with early, but definite OA of the hand, JSW measurements showed that 56% of the patient","[313, 906, 706, 1058]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,7,text,,"[717, 117, 1112, 419]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,body_zone,body_like,empty,False,True +5,8,paragraph_title,"Relationship between changes in joint space, subchondral sclerosis and osteophytes","[717, 456, 1072, 521]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Relationship between changes in joint space, subchondral scl",subsection_heading,0.6,body_zone,heading_like,none,True,True +5,9,text,The results of the studies of OA of the hand $ ^{38} $ and knee $ ^{45} $ showed that the extent of ,"[717, 521, 1114, 1059]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,10,image,,"[324, 1090, 1102, 1539]",media_asset,0.85,media label: image,media_asset,0.85,body_zone,unknown_like,empty,True,True +5,11,figure_title,"Figure 3 Part of a macroradiograph of the metacarpo-phalangeal joints of a patient with hand OA, sho","[316, 1548, 1053, 1601]",figure_caption_candidate,0.92,figure_title label: Figure 3 Part of a macroradiograph of the metacarpo-phalange,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +6,0,header,Quantitative radiography of OA,"[93, 67, 315, 89]",noise,0.9,header label,noise,0.9,tail_nonref_hold_zone,unknown_like,none,False,False +6,1,number,273,"[1061, 68, 1093, 86]",noise,0.9,page number label,noise,0.9,tail_nonref_hold_zone,unknown_like,short_fragment,False,False +6,2,text,"cartilage, measured as joint space narrowing, is a late stage phenomenon of osteoarthritis.","[294, 124, 686, 168]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,body_like,none,True,True +6,3,text,"In the OA hand, the pattern of increased sclerosis and osteophytosis $ ^{39} $ at the joints of the ","[293, 169, 688, 466]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,body_like,none,True,True +6,4,text,"Evaluation of the pattern of joint space narrowing in the OA hand patients, during the study, showed","[292, 465, 686, 959]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,body_like,none,True,True +6,5,text,The results of these investigations show that by using accurate and precise radiographic procedures ,"[290, 957, 685, 1214]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,body_like,none,True,True +6,6,text,"The greater sensitivity of this technique, compared with standard radiography, has improved the chan","[287, 1213, 683, 1614]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +6,7,text,,"[700, 124, 1096, 363]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,tail_nonref_hold_zone,unknown_like,empty,False,True +6,8,paragraph_title,Increasing the accuracy and,"[700, 403, 968, 424]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Increasing the accuracy and",subsection_heading,0.6,tail_nonref_hold_zone,heading_like,none,True,True +6,9,paragraph_title,reproducibility in standard radiography,"[700, 426, 1075, 447]",unknown_structural,0.6,"unnumbered paragraph_title, inferred level subsection_heading: reproducibility in standard radiography",subsection_heading,0.6,tail_nonref_hold_zone,heading_like,none,False,True +6,10,text,"As described above, quantitative microfocal radiography can measure progression, in hand and knee OA","[698, 441, 1093, 917]",body_paragraph,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +6,11,text,Equipment: Accurate measurement within the plain film radiograph is dependent on good spatial resolu,"[696, 917, 1093, 1152]",body_paragraph,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +6,12,text,Patients: Standardisation of the radioanatomical position of the joint is necessary so that the appe,"[695, 1152, 1090, 1428]",body_paragraph,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +6,13,text,Measurement: The boundaries or limits of the radiographic feature to be measured must be defined pre,"[693, 1427, 1089, 1617]",body_paragraph,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +7,0,number,274,"[110, 71, 143, 90]",noise,0.9,page number label,noise,0.9,,unknown_like,short_fragment,False,False +7,1,header,Buckland-Wright,"[986, 71, 1104, 91]",noise,0.9,header label,noise,0.9,,unknown_like,short_fragment,False,False +7,2,text,analysis and overall reduce the time taken for the mensural procedures.,"[310, 127, 699, 173]",body_paragraph,0.6,default body_paragraph for text label; promoted in tail spread from body_paragraph,body_paragraph,0.6,,body_like,none,True,True +7,3,paragraph_title,Conclusion,"[310, 212, 423, 233]",section_heading,0.9,explicit scholarly heading: Conclusion,section_heading,0.9,,heading_like,canonical_section_name,True,True +7,4,text,"To measure OA progression, it is necessary to establish a universally acceptable method for accurate","[310, 235, 701, 705]",body_paragraph,0.6,default body_paragraph for text label; promoted in tail spread from body_paragraph,body_paragraph,0.6,,body_like,none,True,True +7,5,text,I wish to express my gratitude to Dr Charles Hutton for inviting me to write this article and to num,"[312, 721, 702, 827]",body_paragraph,0.6,default body_paragraph for text label; promoted in tail spread from body_paragraph,body_paragraph,0.6,,unknown_like,none,True,True +7,6,reference_content,"1 Resnick D, Niwayama G. Degenerative diseases of extra-spinal locations. In: Resnick D, Niwayama G,","[323, 882, 700, 939]",reference_item,0.85,"reference content label: 1 Resnick D, Niwayama G. Degenerative diseases of extra-spin",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,7,reference_content,"2 Altman R, Asch E, Block D, et al. Development of criteria for the classification and reporting of ","[322, 939, 700, 982]",reference_item,0.85,"reference content label: 2 Altman R, Asch E, Block D, et al. Development of criteria ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,8,reference_content,"3 Altman R, Fries J F, Bloch D A, et al. Radiographic assessment of progression in osteoarthritis. A","[322, 982, 701, 1025]",reference_item,0.85,"reference content label: 3 Altman R, Fries J F, Bloch D A, et al. Radiographic assess",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,9,reference_content,"4 Kallman D A, Wigley F M, Scott W W, Hochberg M C, Tobin J D. New radiographic grading scales for o","[323, 1025, 701, 1067]",reference_item,0.85,"reference content label: 4 Kallman D A, Wigley F M, Scott W W, Hochberg M C, Tobin J ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,10,reference_content,"5 Larsen A. Radiographic evaluation of osteoarthritis in therapeutic trials. In: Verbruggen G, Veys ","[323, 1068, 702, 1125]",reference_item,0.85,reference content label: 5 Larsen A. Radiographic evaluation of osteoarthritis in the,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,11,reference_content,hip. Ann Rheum Dis 1962; 21: 31–9.,"[324, 1124, 701, 1156]",reference_item,0.85,reference content label: hip. Ann Rheum Dis 1962; 21: 31–9.,reference_item,0.85,,unknown_like,none,True,True +7,12,reference_content,7 Ahlback S. Osteoarthritis of the knee: a radiographic investigation. Acta Radiol 1968; (suppl): 1–,"[324, 1153, 701, 1183]",reference_item,0.85,reference content label: 7 Ahlback S. Osteoarthritis of the knee: a radiographic inve,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,13,reference_content,"8 Lequesne M. Clinical features, diagnostic criteria, functional assessments and radiological classi","[324, 1183, 702, 1238]",reference_item,0.85,"reference content label: 8 Lequesne M. Clinical features, diagnostic criteria, functi",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,14,reference_content,"9 Schouten J S A G, van den Ouweland F A, Valkenburg H A. A 12 year follow up study in the general p","[322, 1238, 703, 1295]",reference_item,0.85,"reference content label: 9 Schouten J S A G, van den Ouweland F A, Valkenburg H A. A ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,15,reference_content,"10 Dieppe P, Cushnaghan J, McAlindon T. Epidemiology, clinical course and outcome of knee osteoarthr","[317, 1296, 703, 1364]",reference_item,0.85,"reference content label: 10 Dieppe P, Cushnaghan J, McAlindon T. Epidemiology, clinic",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,16,reference_content,"11 Wevers H W, Siu D, Cooke T D V. A quantitative method of assessing malalignment and joint space l","[318, 1364, 703, 1405]",reference_item,0.85,"reference content label: 11 Wevers H W, Siu D, Cooke T D V. A quantitative method of ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,17,reference_content,"12 Siu D, Cooke T D V, Broekhoven L D, et al. A standardized technique for lower limb radiography, p","[318, 1406, 703, 1446]",reference_item,0.85,"reference content label: 12 Siu D, Cooke T D V, Broekhoven L D, et al. A standardized",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,18,reference_content,"13 Jonson H, Karholm J, Elmqvist L-G. Kinematics of active knee extension after tear of the anterior","[318, 1447, 703, 1488]",reference_item,0.85,"reference content label: 13 Jonson H, Karholm J, Elmqvist L-G. Kinematics of active k",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,19,reference_content,"14 Leach R E, Gregg T, Siber F J. Weight bearing radiography in osteoarthritis of the knee. Radiolog","[320, 1489, 703, 1516]",reference_item,0.85,"reference content label: 14 Leach R E, Gregg T, Siber F J. Weight bearing radiography",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,20,reference_content,15 Menkes C J. Radiographic criteria for classification of OA. § Rheumatol 1991: 18 (suppl 27): 13–5,"[320, 1515, 703, 1543]",reference_item,0.85,reference content label: 15 Menkes C J. Radiographic criteria for classification of O,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,21,reference_content,"16 Fife R S, Brant K D, Braunstein E M, et al. Relationship between arthroscopic evidence of cartila","[320, 1543, 705, 1612]",reference_item,0.85,"reference content label: 16 Fife R S, Brant K D, Braunstein E M, et al. Relationship ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,22,reference_content,"17 Brandt K D, Fife R S, Braunstein E M, Katz B. Radiographic grading of the severity of knee osteoa","[718, 129, 1104, 214]",reference_item,0.85,"reference content label: 17 Brandt K D, Fife R S, Braunstein E M, Katz B. Radiographi",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,23,reference_content,"18 Dacre J E, Huskisson E C. The automatic assessment of knee radiographs in osteoarthritis using di","[719, 215, 1106, 257]",reference_item,0.85,"reference content label: 18 Dacre J E, Huskisson E C. The automatic assessment of kne",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,24,reference_content,"19 Dougados M, Gueguen A, Nguyen M, et al. Longitudinal radiologic evaluation of osteoarthritis of t","[719, 258, 1104, 299]",reference_item,0.85,"reference content label: 19 Dougados M, Gueguen A, Nguyen M, et al. Longitudinal radi",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,25,reference_content,"20 Messieh S S, Fowler P J, Munro T. Anteroposterior radiographs of the osteoarthritic knee. J Bone ","[719, 300, 1104, 342]",reference_item,0.85,"reference content label: 20 Messieh S S, Fowler P J, Munro T. Anteroposterior radiogr",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,26,reference_content,"21 Resnick D, Vint V. The ‘tunnel’ view in assessment of cartilage loss in osteoarthritis of the kne","[719, 342, 1105, 384]",reference_item,0.85,"reference content label: 21 Resnick D, Vint V. The ‘tunnel’ view in assessment of car",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,27,reference_content,"22 Altman R, Fries J F, Block D A, et al. Radiological assessment of progression in osteoarthritis. ","[719, 385, 1105, 427]",reference_item,0.85,"reference content label: 22 Altman R, Fries J F, Block D A, et al. Radiological asses",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,28,reference_content,"23 Adams M E, Wallace C J. Quantitative imaging of osteoarthritis. Semin Arthritis Rheum 1991; 20: 2","[719, 428, 1104, 457]",reference_item,0.85,"reference content label: 23 Adams M E, Wallace C J. Quantitative imaging of osteoarth",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,29,reference_content,"24 Jonsson K, Buckwalter K, Helvie M, Niklason L, Martel W. Precision of hyaline cartilage thickness","[720, 457, 1104, 500]",reference_item,0.85,"reference content label: 24 Jonsson K, Buckwalter K, Helvie M, Niklason L, Martel W. ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,30,reference_content,"25 Dacre J E, Coppock J S, Herbert K E, Perrett D, Huskisson E C. Development of a new radiographic ","[720, 499, 1106, 555]",reference_item,0.85,"reference content label: 25 Dacre J E, Coppock J S, Herbert K E, Perrett D, Huskisson",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,31,reference_content,"26 Jasani M K. Diclofenac and the osteoarthritis disease process in cartilage. In: Moskowitz R, Hiro","[719, 556, 1106, 613]",reference_item,0.85,reference content label: 26 Jasani M K. Diclofenac and the osteoarthritis disease pro,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,32,reference_content,"27 Spector T D, Dacre J E, Harris P A, Huskisson E C. Radiological progression of osteoarthritis: an","[719, 612, 1106, 668]",reference_item,0.85,"reference content label: 27 Spector T D, Dacre J E, Harris P A, Huskisson E C. Radiol",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,33,reference_content,"28 Browne M A, Gaydecki P A, Gough R F, Grennant D M, Khalil S I, Mamtora H. Radiographic image anal","[720, 668, 1106, 726]",reference_item,0.85,"reference content label: 28 Browne M A, Gaydecki P A, Gough R F, Grennant D M, Khalil",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,34,reference_content,"29 Gaydecki P A, Browne M, Mamtora H, Grennant D M. Measurement of radiographic changes occurring in","[719, 727, 1107, 784]",reference_item,0.85,"reference content label: 29 Gaydecki P A, Browne M, Mamtora H, Grennant D M. Measurem",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,35,reference_content,"30 Dacre J E, Scott D L, Da Silva J A P, Welsh G, Huskisson E C. Joint space in radiologically norma","[721, 783, 1106, 826]",reference_item,0.85,"reference content label: 30 Dacre J E, Scott D L, Da Silva J A P, Welsh G, Huskisson ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,36,reference_content,31 Buckland-Wright J C. X-ray assessment of activity in rheumatoid disease. Br J Rheumatol 1983; 22:,"[720, 827, 1107, 856]",reference_item,0.85,reference content label: 31 Buckland-Wright J C. X-ray assessment of activity in rheu,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,37,reference_content,32 Buckland-Wright J C. Microfocal radiographic examination of erosions in the wrist and hand of pat,"[721, 856, 1107, 897]",reference_item,0.85,reference content label: 32 Buckland-Wright J C. Microfocal radiographic examination ,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,38,reference_content,33 Buckland-Wright J C. A new high-definition microfocal x-ray unit. Br J Radiol 1989; 62: 201–8.,"[721, 899, 1107, 925]",reference_item,0.85,reference content label: 33 Buckland-Wright J C. A new high-definition microfocal x-r,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,39,reference_content,"34 Buckland-Wright J C, Bradshaw C R. Clinical applications of high definition microfocal radiograph","[721, 926, 1107, 968]",reference_item,0.85,"reference content label: 34 Buckland-Wright J C, Bradshaw C R. Clinical applications ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,40,reference_content,"35 Buckland-Wright J C. Carmichael I, Walker S R. Quantitative microfocal radiography accurately det","[721, 968, 1108, 1024]",reference_item,0.85,"reference content label: 35 Buckland-Wright J C. Carmichael I, Walker S R. Quantitati",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,41,reference_content,36 Clarke G S. Quantitative microfocal radiographic assessment of changes in the joint structure of ,"[721, 1026, 1108, 1068]",reference_item,0.85,reference content label: 36 Clarke G S. Quantitative microfocal radiographic assessme,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,42,reference_content,"37 Foley-Nolan D, Stack J P, Ryan M, et al. Magnetic resonance imaging in the assessment of rheumato","[721, 1068, 1108, 1124]",reference_item,0.85,"reference content label: 37 Foley-Nolan D, Stack J P, Ryan M, et al. Magnetic resonan",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,43,reference_content,"38 Buckland-Wright J C. Macfarlane D G, Lynch J A, Clark B. Quantitative microfocal radiographic ass","[721, 1124, 1108, 1181]",reference_item,0.85,"reference content label: 38 Buckland-Wright J C. Macfarlane D G, Lynch J A, Clark B. ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,44,reference_content,"39 Buckland-Wright J C, Macfarlane D G, Lynch J. Relationship between joint space width and subchond","[721, 1181, 1108, 1238]",reference_item,0.85,"reference content label: 39 Buckland-Wright J C, Macfarlane D G, Lynch J. Relationshi",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,45,reference_content,"40 Buckland-Wright J C, Macfarlane D G, Fogelman I, Emery P, Lynch J A. Technetium 99mm methylene di","[721, 1238, 1109, 1294]",reference_item,0.85,"reference content label: 40 Buckland-Wright J C, Macfarlane D G, Fogelman I, Emery P,",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,46,reference_content,"41 Macfarlane D G, Buckland-Wright J C, Emery P, Fogelman I, Lynch J. Comparison of clinical, radion","[721, 1294, 1108, 1351]",reference_item,0.85,"reference content label: 41 Macfarlane D G, Buckland-Wright J C, Emery P, Fogelman I,",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,47,reference_content,"42 Buckland-Wright J C, Macfarlane D G, Lynch J A. Osteophytes in the arthritic hand: their incidenc","[722, 1350, 1108, 1404]",reference_item,0.85,"reference content label: 42 Buckland-Wright J C, Macfarlane D G, Lynch J A. Osteophyt",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,48,reference_content,"43 Kellgren J H, Lawrence J S. Radiological assessment of osteoarthrosis. Ann Rheum Dis 1957; 16: 49","[722, 1405, 1109, 1433]",reference_item,0.85,"reference content label: 43 Kellgren J H, Lawrence J S. Radiological assessment of os",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,49,reference_content,44 Moll J M H. Investigation of osteoarthritis. Clin Rheum Dis 1977; 2: 587–613.,"[722, 1432, 1109, 1460]",reference_item,0.85,reference content label: 44 Moll J M H. Investigation of osteoarthritis. Clin Rheum D,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,50,reference_content,"45 Buckland-Wright J C, Macfarlane D G, Lynch J A, Jasani M K. Measurement of joint space loss in os","[722, 1459, 1109, 1540]",reference_item,0.85,"reference content label: 45 Buckland-Wright J C, Macfarlane D G, Lynch J A, Jasani M ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +7,51,reference_content,"46 Buckland-Wright J C, Macfarlane D G, Jasani M K, Lynch J A. Quantitative microfocal radiographic ","[724, 1540, 1111, 1610]",reference_item,0.85,"reference content label: 46 Buckland-Wright J C, Macfarlane D G, Jasani M K, Lynch J ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,0,number,275,"[100, 62, 133, 79]",noise,0.9,page number label,noise,0.9,,unknown_like,short_fragment,False,False +8,1,header,Buckland-Wright,"[982, 62, 1101, 80]",noise,0.9,header label,noise,0.9,,unknown_like,short_fragment,False,False +8,2,reference_content,"47 Buckland-Wright J C, Macfarlane D G, Jasani M K, Lynch J A. Changes in OA knee joint space width ","[304, 114, 693, 186]",reference_item,0.85,"reference content label: 47 Buckland-Wright J C, Macfarlane D G, Jasani M K, Lynch J ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,3,reference_content,"48 Buckland-Wright J C. Macfarlane D G, Jasani M K, Lynch J A. Joint space width measures cartilage ","[305, 188, 693, 258]",reference_item,0.85,"reference content label: 48 Buckland-Wright J C. Macfarlane D G, Jasani M K, Lynch J ",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,4,reference_content,49 Maroudas A. Balance between swelling pressure and collagen tension in normal and degenerative car,"[306, 259, 692, 301]",reference_item,0.85,reference content label: 49 Maroudas A. Balance between swelling pressure and collage,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,5,reference_content,"50 Mankin H J, Thrasher A Z. Water content and binding in normal and osteoarthritic human cartilage.","[306, 301, 693, 343]",reference_item,0.85,"reference content label: 50 Mankin H J, Thrasher A Z. Water content and binding in no",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,6,reference_content,"51 Mow V C, Setton L A, Ratcliff A, Howell D S, Buckwalter J A. Structure-function relationships of ","[305, 343, 692, 427]",reference_item,0.85,"reference content label: 51 Mow V C, Setton L A, Ratcliff A, Howell D S, Buckwalter J",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,7,reference_content,"52 Lane L B, Villacin A, Bullough P G. The vascularity and remodelling of subchondral bone and calci","[306, 428, 692, 484]",reference_item,0.85,"reference content label: 52 Lane L B, Villacin A, Bullough P G. The vascularity and r",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,8,reference_content,"53 Bullough P G, Goodfellow J W. Incongruent surfaces in the hip joint. Nature 1968; 217: 1290.","[305, 485, 692, 515]",reference_item,0.85,"reference content label: 53 Bullough P G, Goodfellow J W. Incongruent surfaces in the",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,9,reference_content,"54 McDevitt C A, Gilbertson E, Muir H. An experimental model of osteoarthritis: early morphological ","[305, 514, 692, 570]",reference_item,0.85,"reference content label: 54 McDevitt C A, Gilbertson E, Muir H. An experimental model",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,10,reference_content,"55 Mankin H J, Brant K D. Biochemistry and metabolism of cartilage in osteoarthritis. In: Moscowitz ","[304, 571, 692, 641]",reference_item,0.85,"reference content label: 55 Mankin H J, Brant K D. Biochemistry and metabolism of car",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,11,reference_content,56 Fassbender H G. Significance of endogenous and exogenous mechanisms in the development of osteoar,"[304, 643, 691, 726]",reference_item,0.85,reference content label: 56 Fassbender H G. Significance of endogenous and exogenous ,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,12,reference_content,57 Lanyon C E. Functional strain as a determinant for bone remodelling. Calc Tissue Int 1984; 36: S5,"[305, 726, 690, 756]",reference_item,0.85,reference content label: 57 Lanyon C E. Functional strain as a determinant for bone r,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,13,reference_content,"58 Williams J M, Brandt K D. Exercise increases osteophyte","[305, 755, 691, 770]",reference_item,0.85,"reference content label: 58 Williams J M, Brandt K D. Exercise increases osteophyte",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,14,reference_content,formation and diminishes fibrillation following chemically induced articular cartilage injury. § Ana,"[746, 116, 1099, 158]",reference_item,0.85,reference content label: formation and diminishes fibrillation following chemically i,reference_item,0.85,,unknown_like,none,True,True +8,15,reference_content,59 Gilbertson E M M. Development of periarticular osteophytes in experimentally induced osteoarthrit,"[714, 158, 1099, 202]",reference_item,0.85,reference content label: 59 Gilbertson E M M. Development of periarticular osteophyte,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,16,reference_content,60 Napier J R. The form and function of the carpo-metacarpal joint of the thumb. $ \mathcal{F} $ An,"[713, 202, 1100, 232]",reference_item,0.85,reference content label: 60 Napier J R. The form and function of the carpo-metacarpal,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,17,reference_content,"61 Backhouse K M, Hutchings R T. A colour atlas of surface anatomy, clinical and applied. Netherland","[714, 232, 1099, 272]",reference_item,0.85,"reference content label: 61 Backhouse K M, Hutchings R T. A colour atlas of surface a",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,18,reference_content,"62 Tubiana R, Thomine J-M, Mackin E. Examination of hand and upper limb. Philadelphia: Saunders, 198","[713, 273, 1100, 302]",reference_item,0.85,"reference content label: 62 Tubiana R, Thomine J-M, Mackin E. Examination of hand and",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,19,reference_content,"63 Jones A R, Unsworth A, Haslock I. A microcomputer controlled hand assessment system used for clin","[712, 302, 1098, 344]",reference_item,0.85,"reference content label: 63 Jones A R, Unsworth A, Haslock I. A microcomputer control",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,20,reference_content,"64 Moskowitz R W. Experimental models of osteoarthritis. In: Moskowitz R W, Howell D S, Goldberg V M","[712, 345, 1099, 400]",reference_item,0.85,reference content label: 64 Moskowitz R W. Experimental models of osteoarthritis. In:,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,21,reference_content,"65 Radin E L, Paul I L, Rose R M. Role of mechanical factors in pathogenesis of primary osteoarthrit","[712, 401, 1098, 443]",reference_item,0.85,"reference content label: 65 Radin E L, Paul I L, Rose R M. Role of mechanical factors",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,22,reference_content,66 Buckland-Wright J C. The early lesion in subchondral bone in osteoarthritis: a microfocal radiogr,"[711, 444, 1099, 527]",reference_item,0.85,reference content label: 66 Buckland-Wright J C. The early lesion in subchondral bone,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,23,reference_content,67 Sokoloff L. Loading and motion in relation to ageing and degeneration of joints: implications for,"[711, 529, 1098, 613]",reference_item,0.85,reference content label: 67 Sokoloff L. Loading and motion in relation to ageing and ,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,24,reference_content,68 Stecher R M. Heberden's nodes. A clinical description of osteoarthritis of the finger joints. Ann,"[711, 614, 1098, 656]",reference_item,0.85,reference content label: 68 Stecher R M. Heberden's nodes. A clinical description of ,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,25,reference_content,"69 Buckland-Wright J C. Imaging and measurement of change in osteoarthritis. In: Barrowclough D, ed.","[710, 658, 1099, 714]",reference_item,0.85,reference content label: 69 Buckland-Wright J C. Imaging and measurement of change in,reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True +8,26,reference_content,"70 Lynch J A, Buckland-Wright J C, Hawkes D J. Automated measurement of interbone distance on macror","[711, 714, 1099, 771]",reference_item,0.85,"reference content label: 70 Lynch J A, Buckland-Wright J C, Hawkes D J. Automated mea",reference_item,0.85,reference_zone,reference_like,reference_numeric_dot,True,True diff --git a/tests/fixtures/ocr_real_papers/CAQNW9Q2/expectations.json b/tests/fixtures/ocr_real_papers/CAQNW9Q2/expectations.json index dcf6de84..e5fb5fcc 100644 --- a/tests/fixtures/ocr_real_papers/CAQNW9Q2/expectations.json +++ b/tests/fixtures/ocr_real_papers/CAQNW9Q2/expectations.json @@ -1,26 +1,151 @@ { "document": { - "expected_reader_figure_count_min": 2, + "expected_reader_figure_count_min": 3, "expected_reference_zone": true, - "expected_abstract_span": true + "expected_abstract_span": false, + "notes": [ + "IDEAL expectations for CAQNW9Q2 (Buckland-Wright 1994 review article).", + "No abstract. Review article in Annals of the Rheumatic Diseases 1994; 53: 268-275.", + "Page 1: header/journal line, REVIEW label, title, author, body text, section heading.", + "Figures 1-3 are radiographic X-ray images - must be reader-visible.", + "References span pages 7-8 (refs 1-70).", + "Headings: raw_label=paragraph_title, final role=section_heading/subsection_heading.", + "Running headers: noise/frontmatter_noise, must_not_render.", + "Title/authors: frontmatter artifact, not body_zone." + ] }, "pages": { "1": { - "expected_roles": [ - {"block_pattern": "Quantitative radiography", "role": "title_or_doc_title", "confidence": "medium"} + "assertions": [ + {"text_contains": "Annals of the Rheumatic Diseases", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true, "notes": "Real: noise - acceptable, just needs must_not_render"}, + {"text_equals": "REVIEW", "expected_role": "frontmatter_noise", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: unknown_structural - should be frontmatter_noise (article type label, not body content)"}, + {"text_contains": "Quantitative radiography of osteoarthritis", "expected_role": "paper_title", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: unknown_structural - should be paper_title"}, + {"text_contains": "J C Buckland-Wright", "expected_role": "authors", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: unknown_structural - should be authors"}, + {"text_contains": "Radiography is important", "expected_role": "body_paragraph", "expected_zone": "body_zone", "notes": "Real: body_paragraph in frontmatter_main_zone - zone is wrong"}, + {"text_equals": "Standard radiography", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone", "notes": "Real: unknown_structural - should be heading"}, + {"text_contains": "Correspondence to:", "expected_role": "frontmatter_support", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: NOT FOUND - selector was wrong, actual text is 'Correspondence to: Dr J C Buckland-Wright'"} + ] + }, + "2": { + "assertions": [ + {"text_equals": "Quantitative radiography of OA", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true, "notes": "Running header - should not render"}, + {"text_equals": "Quantitative standard radiography", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone", "notes": "Real: subsection_heading - correct!"} ], - "expected_non_body": [0, 1, 4, 5], - "expected_reference_rules": [ - {"block_id_comment": "no block on page 1 should be reference_item"} + "expected_object_ownership": [ + { + "object_type": "figure", + "figure_number": 1, + "caption_text_startswith": "Figure 1 Standard radiographic appearance", + "must_not_be_status": "LEGEND_ONLY", + "must_have_asset_or_candidate_asset": true + } + ] + }, + "3": { + "assertions": [ + {"text_equals": "Buckland-Wright", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true, "notes": "Running header"}, + {"text_equals": "Microfocal radiography", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone"} + ] + }, + "4": { + "assertions": [ + {"text_equals": "Quantitative radiography of OA", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true}, + {"text_equals": "Standardisation of macroradiographic procedure", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone"} + ], + "expected_object_ownership": [ + { + "object_type": "figure", + "figure_number": 2, + "must_not_be_status": "LEGEND_ONLY", + "must_have_asset_or_candidate_asset": true + } + ] + }, + "5": { + "assertions": [ + {"text_equals": "Buckland-Wright", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true}, + {"text_equals": "Quantitative microfocal radiography", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone"}, + {"text_equals": "Joint space width", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone"}, + {"text_contains": "Relationship between changes in joint space", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone"} + ], + "expected_object_ownership": [ + { + "object_type": "figure", + "figure_number": 3, + "must_not_be_status": "LEGEND_ONLY", + "must_have_asset_or_candidate_asset": true + } + ] + }, + "6": { + "assertions": [ + {"text_equals": "Quantitative radiography of OA", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true}, + {"text_contains": "cartilage, measured as joint space narrowing", "expected_role": "body_paragraph", "expected_zone": "body_zone", "notes": "Real: body_paragraph in tail_nonref_hold_zone - zone is wrong, this is body content"}, + {"type": "merged_heading", "parts": ["Increasing the accuracy and", "reproducibility in standard radiography"], "expected_text": "Increasing the accuracy and reproducibility in standard radiography", "expected_role_any_of": ["section_heading", "subsection_heading"], "notes": "Real: split into 2 blocks, heading merge artifact needed"} ] }, "7": { - "expected_order_relations": [ - {"before_text": "Conclusion", "after_text": "References", "layer": "render_order_markdown"} - ], - "expected_reference_rules": [ - {"must_not_render_references_as_body": true} + "assertions": [ + {"text_equals": "Buckland-Wright", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true}, + {"text_equals": "Conclusion", "expected_role_any_of": ["section_heading", "subsection_heading"], "expected_zone": "body_zone"}, + {"text_contains": "I wish to express my gratitude", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone", "notes": "No Acknowledgements heading - just backmatter body text"} + ] + }, + "8": { + "assertions": [ + {"text_equals": "Buckland-Wright", "expected_role_any_of": ["noise", "frontmatter_noise"], "must_not_render": true, "must_not_index": true}, + {"text_contains": "formation and diminishes fibrillation", "expected_role": "reference_item", "expected_zone": "reference_zone", "notes": "Real: unknown_structural in empty zone - should be reference_item (ref 58 continuation)"} ] } - } + }, + "expected_bugs": [ + { + "bug": "title_and_author_not_classified", + "pages": [1], + "description": "Paper title gets role=unknown_structural (seed_role=paper_title but gate rejected). Author name gets role=unknown_structural. Both should be paper_title/authors.", + "fix": "paper_title/authors should be accepted by source-backed frontmatter artifact gate" + }, + { + "bug": "page1_body_zone_eaten_by_frontmatter", + "pages": [1], + "description": "Page 1 body text (Radiography is important..., Scoring systems...) has zone=frontmatter_main_zone instead of body_zone. Frontmatter zone has no lower boundary.", + "fix": "Page 1 frontmatter zone must have lower boundary, not swallow two-column body text" + }, + { + "bug": "running_headers_not_render_filtered", + "pages": [2, 3, 4, 5, 6, 7, 8], + "description": "Running headers (Buckland-Wright, Quantitative radiography of OA) have role=noise but some still render. Need must_not_render gate.", + "fix": "Running headers should be noise with must_not_render=true" + }, + { + "bug": "page6_tail_misclassification", + "pages": [6], + "description": "Page 6 body text (cartilage, measured as joint space narrowing...) is in tail_nonref_hold_zone instead of body_zone. Tail detection spread too early.", + "fix": "tail_nonref_hold_zone spread must not consume body continuation pages" + }, + { + "bug": "page6_heading_split_no_merge", + "pages": [6], + "description": "Heading 'Increasing the accuracy and reproducibility in standard radiography' is split into 2 blocks with no merge artifact.", + "fix": "Heading merge logic needed for split headings" + }, + { + "bug": "reference_continuation_lost", + "pages": [8], + "description": "Reference 58 continuation (formation and diminishes fibrillation) has role=unknown_structural in empty zone instead of reference_item in reference_zone.", + "fix": "Reference continuation detection needs improvement" + }, + { + "bug": "review_label_not_classified", + "pages": [1], + "description": "REVIEW label has role=unknown_structural instead of frontmatter_noise.", + "fix": "Article type labels need dedicated classification as frontmatter_noise" + }, + { + "bug": "figure_captions_not_tested", + "pages": [2, 4, 5], + "description": "Figure object ownership not tested in expectations. Need to verify figures 1-3 are reader-visible with captions.", + "fix": "Add figure object ownership assertions to expectations" + } + ] } diff --git a/tests/fixtures/ocr_real_papers/DWQQK2YB/block_trace.csv b/tests/fixtures/ocr_real_papers/DWQQK2YB/block_trace.csv new file mode 100644 index 00000000..d0b9966d --- /dev/null +++ b/tests/fixtures/ocr_real_papers/DWQQK2YB/block_trace.csv @@ -0,0 +1,288 @@ +page,block_id,raw_label,content_preview,bbox,role,role_confidence,evidence,seed_role,seed_confidence,zone,style_family,marker_type,render_default,index_default +1,0,paragraph_title,Journal Pre-proof,"[190, 206, 475, 246]",frontmatter_noise,0.98,"journal pre-proof marker: page 1, paragraph_title, y=206/1584, width=285/1224",frontmatter_noise,0.98,preproof_cover_zone,heading_like,preproof_marker,False,False +1,1,text,Magnetoresponsive Stem Cell Spheroid-based Cartilage Recovery Platform Utilizing Electromagnetic Fie,"[189, 314, 784, 365]",frontmatter_noise,0.85,page-1 title fallback after pre-proof: y=314/1584,paper_title,0.85,frontmatter_main_zone,support_like,none,False,False +1,2,text,"Ami Yoo, Gwangjun Go, Kim Tien Nguyen, Kyungmin Lee, Hyun-Ki Min, Byungjeon Kang, Chang-Sei Kim, Jiw","[190, 391, 783, 467]",frontmatter_noise,0.8,"page-1 zone author_zone: Ami Yoo, Gwangjun Go, Kim Tien Nguyen, Kyungmin Lee, Hyun-Ki",authors,0.8,frontmatter_main_zone,support_like,none,False,False +1,3,image,,"[846, 209, 1015, 436]",frontmatter_noise,0.85,media label: image,media_asset,0.85,frontmatter_main_zone,support_like,empty,False,False +1,4,text,PII: S0925-4005(19)31768-X,"[190, 511, 632, 536]",frontmatter_noise,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,False,False +1,5,text,DOI: https://doi.org/10.1016/j.snb.2019.127569,"[190, 548, 778, 574]",frontmatter_noise,0.8,page-1 zone journal_furniture_zone: DOI: https://doi.org/10.1016/j.snb.2019.127569,frontmatter_noise,0.8,,unknown_like,none,False,False +1,6,text,Reference: SNB 127569,"[191, 586, 526, 612]",frontmatter_noise,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,False,False +1,7,text,To appear in: Sensors and Actuators: B. Chemical,"[190, 651, 735, 676]",frontmatter_noise,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,False,False +1,8,text,Received Date: 13 August 2019,"[190, 713, 553, 739]",frontmatter_noise,0.7,frontmatter noise text: Received Date: 13 August 2019,frontmatter_noise,0.7,frontmatter_main_zone,support_like,none,False,False +1,9,text,Revised Date: 19 November 2019,"[190, 751, 582, 777]",frontmatter_noise,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,False,False +1,10,text,Accepted Date: 10 December 2019,"[190, 789, 583, 814]",frontmatter_noise,0.8,page-1 zone journal_furniture_zone: Accepted Date: 10 December 2019,frontmatter_noise,0.8,frontmatter_main_zone,support_like,none,False,False +1,11,text,"Please cite this article as: Yoo A, Go G, Nguyen KT, Lee K, Min H-Ki, Kang B, Kim C-Sei, Han J, Park","[189, 876, 1019, 976]",frontmatter_noise,0.8,"page-1 zone journal_furniture_zone: Please cite this article as: Yoo A, Go G, Nguyen KT, Lee K, ",frontmatter_noise,0.8,,unknown_like,none,False,False +1,12,text,"This is a PDF file of an article that has undergone enhancements after acceptance, such as the addit","[188, 1019, 1004, 1188]",frontmatter_noise,0.6,default body_paragraph for text label,body_paragraph,0.6,frontmatter_main_zone,support_like,none,False,False +1,13,text,© 2019 Published by Elsevier.,"[189, 1209, 461, 1234]",frontmatter_noise,0.8,page-1 zone journal_furniture_zone: © 2019 Published by Elsevier.,frontmatter_noise,0.8,frontmatter_main_zone,support_like,none,False,False +2,0,header,Journal Pre-proof,"[502, 1, 832, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p2 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +2,1,text,TITLE,"[76, 101, 146, 128]",unknown_structural,0.3,"short text, uncertain role",unknown_structural,0.3,frontmatter_side_zone,support_like,short_fragment,False,True +2,2,doc_title,Magnetoresponsive Stem Cell Spheroid-based Cartilage Recovery Platform Utilizing Electromagnetic Fie,"[74, 155, 1028, 268]",unknown_structural,0.2,unrecognized label 'doc_title',unknown_structural,0.2,body_zone,body_like,none,False,True +2,3,text,"Ami Yooa, †, Gwangjun Goa, b, †, Kim Tien Nguyena, ","[74, 355, 1026, 445]",frontmatter_support,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +2,4,text,"a. Korea Institute of Medical Microrobotics, 43-26 Cheomdangwagi-ro, Buk-gu, Gwangju, 61011, Korea","[82, 535, 1028, 621]",frontmatter_support,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +2,5,text,"b. School of Mechanical Engineering, Chonnam National University, 77 Yongbong-ro, Buk-gu, Gwangju, 6","[81, 645, 1028, 732]",frontmatter_support,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +2,6,text,These authors contributed equally to this work.,"[75, 811, 548, 843]",frontmatter_noise,0.6,default body_paragraph for text label; frontmatter_side_zone excluded from body flow,frontmatter_noise,0.6,frontmatter_side_zone,support_like,none,False,False +2,7,text,"*Corresponding author: judyvet@jnu.ac.kr (J.H.), jop@jnu.ac.kr (J.-O.P.), and eunpyochoi@jnu.ac.kr (","[74, 864, 1032, 953]",frontmatter_noise,0.6,default body_paragraph for text label; frontmatter_side_zone excluded from body flow,frontmatter_noise,0.6,frontmatter_side_zone,support_like,none,False,False +2,8,paragraph_title,Highlights,"[75, 1115, 188, 1143]",structured_insert,0.6,"unnumbered paragraph_title, inferred level sub_subsection_heading: Highlights",sub_subsection_heading,0.6,body_zone,heading_like,short_fragment,False,False +2,9,text,- We propose the use of a magnetoresponsive stem cell spheroid (MR-SCS)-based cartilage recovery pla,"[109, 1144, 1014, 1253]",structured_insert,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,False,False +2,10,text,- Locomotion of MR-SCS that was mediated by the electromagnetic actuationsystem was successfully dem,"[111, 1278, 966, 1332]",structured_insert,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,False,False +2,11,text,- Low-frequency electromagnetic field stimulation of MR-SCS resulted in increased expression levels ,"[110, 1359, 1016, 1414]",structured_insert,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,False,False +3,0,header,Journal Pre-proof,"[503, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p3 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +3,1,abstract,• Histological evaluation revealed an apparent improvement in the regeneration of cartilage tissue i,"[112, 101, 1015, 184]",abstract_body,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +3,2,paragraph_title,Abstract,"[74, 262, 191, 297]",abstract_heading,0.95,abstract heading,abstract_heading,0.95,body_zone,heading_like,short_fragment,True,True +3,3,abstract,Mesenchymal stem cells (MSCs) provide a promising source for cartilage tissue regeneration strategie,"[71, 319, 1033, 1245]",abstract_body,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +3,4,text,"Keywords: stem cell, cell spheroid, electromagnetic actuation, electromagnetic field stimulation","[75, 1321, 1011, 1352]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +3,5,paragraph_title,1. Introduction,"[77, 1434, 264, 1465]",section_heading,0.85,paragraph_title label with numbering: 1. Introduction,section_heading,0.85,body_zone,heading_like,heading_numbered,True,True +4,0,paragraph_title,Journal Pre-proof,"[503, 1, 832, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p4 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +4,1,text,"As the world population increases in age and in life expectancy, joint related disorders such as ost","[74, 99, 1033, 1351]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +4,2,text,"MSCs provide a promising cell source for cartilage repair due to their self-renewal ability, high pr","[74, 1369, 1032, 1457]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,0,header,Journal Pre-proof,"[504, 1, 831, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p5 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +5,1,text,tissues and organs [9]. A number of studies have investigated the applicability of MSCs in regenerat,"[74, 101, 1033, 524]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,2,text,"In an effort to enhance the targeting efficiency of stem cells, several studies have suggested the u","[74, 543, 1032, 1186]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +5,3,text,"Given this, the proof of concept of a three-dimensional (3D) scaffold-based stem cell delivery syste","[75, 1205, 1032, 1456]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +6,0,header,Journal Pre-proof,"[504, 1, 832, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p6 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +6,1,text,still some unsolved issues associated with long-term safety and efficacy related to these materials.,"[75, 100, 1032, 687]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +6,2,text,"Currently, stem cell-based tissue engineering strategies that incorporate biochemical and biophysica","[75, 709, 1033, 1237]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +6,3,text,"To accomplish this, we propose the use of a magnetoresponsive stem cell spheroid-based cartilage rec","[75, 1261, 1032, 1456]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +7,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p7 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +7,1,text,hundred microns by co-culturing the mesenchymal stem cell (MSC) with magnetic nanoparticles (MNPs) v,"[75, 101, 1032, 574]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +7,2,paragraph_title,2. Experimental section,"[75, 664, 365, 694]",section_heading,0.85,paragraph_title label with numbering: 2. Experimental section,section_heading,0.85,,reference_like,reference_numeric_dot,True,True +7,3,paragraph_title,2.1 Cell culture,"[76, 726, 241, 754]",subsection_heading,0.85,paragraph_title label with numbering: 2.1 Cell culture,subsection_heading,0.85,body_zone,heading_like,heading_numbered,True,True +7,4,text,"Mouse mesenchymal stem cells (MSCs) were obtained from American Type Culture Collection (ATCC, VA, U","[74, 782, 1032, 1200]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +7,5,paragraph_title,2.2 Fabrication and characteristics of MNP labeled MR-SCS,"[75, 1279, 705, 1305]",subsection_heading,0.85,paragraph_title label with numbering: 2.2 Fabrication and characteristics of MNP labeled MR-SCS,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +7,6,text,Confluent mouse MSCs were incubated in culture medium containing different concentrations of MNPs (f,"[74, 1334, 1032, 1475]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +8,0,header,Journal Pre-proof,"[504, 1, 830, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p8 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +8,1,text,MNPs. The fluidMAG-D is composed of magnetite core and starch surface coating and prepared as a conc,"[74, 101, 1031, 905]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +8,2,paragraph_title,2.3 Chondrogenic differentiation of the MR-SCS,"[75, 984, 582, 1012]",subsection_heading,0.85,paragraph_title label with numbering: 2.3 Chondrogenic differentiation of the MR-SCS,subsection_heading,0.85,body_zone,heading_like,heading_numbered,True,True +8,3,text,The fabricated MR-SCSs were cultured for 3 weeks in chondrogenic differentiation medium composed of ,"[75, 1040, 1033, 1290]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +8,4,paragraph_title,2.4 Experimental setup for electromagnetic actuation of the MR-SCS,"[74, 1371, 791, 1398]",subsection_heading,0.85,paragraph_title label with numbering: 2.4 Experimental setup for electromagnetic actuation of the ,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +9,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p9 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +9,1,text,The electromagnetic actuation of the MR-SCS was conducted using the EMA system that consists of eigh,"[74, 100, 1032, 685]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +9,2,text,"For the 3D locomotion test of the MR-SCS using the EMA system, a single MR-SCS was placed in the acr","[74, 708, 1032, 962]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +9,3,paragraph_title,2.5 Experimental setup for low-field electromagnetic stimulation of the MR-SCS,"[74, 1039, 907, 1068]",subsection_heading,0.85,paragraph_title label with numbering: 2.5 Experimental setup for low-field electromagnetic stimula,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +9,4,text,"For electromagnetic stimulation of the MR-SCS, an LF-EMF stimulation system that consists of two squ","[75, 1095, 1032, 1454]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +10,0,header,Journal Pre-proof,"[505, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p10 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +10,1,text,"differentiation of MR-SCS, osteogenic and chondrogenic differentiation of MSCs occurs at the same ma","[75, 102, 1032, 574]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +10,2,paragraph_title,2.6 Ex vivo cartilage defect model and implantation of the MR-SCS,"[74, 653, 771, 681]",subsection_heading,0.85,paragraph_title label with numbering: 2.6 Ex vivo cartilage defect model and implantation of the M,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +10,3,text,Osteochondral discs (8 mm × 8 mm) were harvested from the medial condyles of porcine femurs using an,"[74, 709, 1032, 1126]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +10,4,paragraph_title,2.7 Gene expression analysis,"[76, 1206, 376, 1233]",subsection_heading,0.85,paragraph_title label with numbering: 2.7 Gene expression analysis,subsection_heading,0.85,body_zone,heading_like,heading_numbered,True,True +10,5,text,"To analyze the chondrogenic differentiation of the MR-SCS, relative gene expression changes of chond","[75, 1261, 1031, 1455]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +11,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p11 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +11,1,text,time polymerase chain reaction (PCR) was performed using 5 x HOT FIREPol® EvaGreen® qPCR Mix Plus (R,"[75, 102, 1032, 577]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +11,2,paragraph_title,2.8 Histological and immunohistochemical staining analysis,"[75, 653, 694, 682]",subsection_heading,0.85,paragraph_title label with numbering: 2.8 Histological and immunohistochemical staining analysis,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +11,3,text,"For the histological and immunohistochemical analyses, the MR-SCSs were fixed in 4% formaldehyde for","[75, 709, 1032, 1015]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +11,4,text,"For immunohistochemical staining, cryosectioned MR-SCSs were incubated in Anti-COLIIA1 antibody (Abc","[75, 1039, 1032, 1347]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +11,5,paragraph_title,2.9 Statistical data analysis,"[76, 1426, 360, 1453]",subsection_heading,0.85,paragraph_title label with numbering: 2.9 Statistical data analysis,subsection_heading,0.85,body_zone,heading_like,heading_numbered,True,True +12,0,header,Journal Pre-proof,"[505, 2, 830, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p12 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +12,1,text,All experiments were replicated 3 times and statistical analyses of the in vitro and ex vivo experim,"[75, 101, 1031, 242]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +12,2,paragraph_title,3. Results,"[76, 323, 200, 351]",section_heading,0.85,paragraph_title label with numbering: 3. Results,section_heading,0.85,body_zone,heading_like,heading_numbered,True,True +12,3,paragraph_title,3.1 Morphology and cell viability of the MR-SCS,"[75, 386, 585, 414]",subsection_heading,0.85,paragraph_title label with numbering: 3.1 Morphology and cell viability of the MR-SCS,subsection_heading,0.85,body_zone,heading_like,heading_numbered,True,True +12,4,text,Mouse MSCs and ultra-low attachment surface 96-well plates were used to fabricate the magnetorespons,"[74, 443, 1031, 858]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +12,5,text,The effect of MNPs on the viability of mouse MSCs in 2D and 3D culture was analyzed using the Alamar,"[74, 884, 1032, 1467]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +13,0,paragraph_title,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p13 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +13,1,text,cell/well and after 24 h about $ 2.4 \times 10^4 $ cells/well formed in single 3D spheroid (fig. S3,"[75, 99, 1032, 961]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +13,2,text,"Additionally, as shown in Live/Dead cell images, a higher amount of green fluorescent cells (live ce","[75, 985, 1033, 1457]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +14,0,header,Journal Pre-proof,"[505, 2, 829, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p14 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +14,1,paragraph_title,3.2 Effect of MNPs on chondrogenic differentiation of the MR-SCS,"[74, 157, 771, 183]",subsection_heading,0.85,paragraph_title label with numbering: 3.2 Effect of MNPs on chondrogenic differentiation of the MR,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +14,2,text,"To evaluate the chondrogenic differentiation of the MR-SCS, Alcian blue staining, immunohistochemica","[75, 214, 1032, 794]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +14,3,paragraph_title,3.3 Manipulation of the MR-SCS using EMA system,"[75, 875, 624, 903]",subsection_heading,0.85,paragraph_title label with numbering: 3.3 Manipulation of the MR-SCS using EMA system,subsection_heading,0.85,body_zone,heading_like,heading_numbered,True,True +14,4,text,"The electromagnetic actuation of the MR-SCS was performed using the EMA system, as this system can p","[75, 931, 1032, 1456]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +15,0,header,Journal Pre-proof,"[503, 1, 831, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p15 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +15,1,abstract,linearly increased at fixed magnetic field (40 mT) and changing gradient magnetic fields (0.9 - 1.8 ,"[70, 92, 1039, 1543]",body_paragraph,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +16,0,paragraph_title,Journal Pre-proof,"[504, 1, 832, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p16 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +16,1,abstract,magnetic chain according to the magnetic field generated by the EMA system. The aggregated MR-SCSs w,"[74, 98, 1034, 355]",body_paragraph,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +16,2,abstract,"In an in vivo and clinical tests, the general optical imaging device is not easy to be used for real","[74, 377, 1034, 1463]",body_paragraph,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +17,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p17 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +17,1,abstract,"control and arthroscope imaging to the defect site (Fig. 3h, fig. S13, and video S4, Supporting Info","[74, 99, 1034, 298]",body_paragraph,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +17,2,abstract,Although the proposed EMA system demonstrated the precise 3D targeting delivery of a small number of,"[75, 321, 1033, 1352]",body_paragraph,0.85,abstract label from Paddle OCR,abstract_body,0.85,body_zone,body_like,none,True,True +17,3,paragraph_title,3.4 Effects of LF-EMF stimulation on in vitro chondrogenic differentiation of the MR-SCS,"[74, 1424, 1009, 1454]",subsection_heading,0.85,paragraph_title label with numbering: 3.4 Effects of LF-EMF stimulation on in vitro chondrogenic d,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +18,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p18 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +18,1,text,"To evaluate the effect of LF-EMF stimulation on chondrogenic differentiation of the MR-SCS, immunohi","[75, 101, 1032, 740]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +18,2,paragraph_title,3.5 Effects of LF-EMF stimulation on ex vivo cartilage regeneration of MR-SCS,"[74, 819, 900, 846]",subsection_heading,0.85,paragraph_title label with numbering: 3.5 Effects of LF-EMF stimulation on ex vivo cartilage regen,subsection_heading,0.85,body_zone,body_like,heading_numbered,True,True +18,3,text,We further evaluated the effects of LF-EMF stimulation on ex vivo cartilage regeneration of the MR-S,"[75, 874, 1032, 1180]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +18,4,text,"After 8 weeks of cultivation with the MR-SCSs, the defective areas of the discs were filled with tis","[75, 1207, 1032, 1455]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +19,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p19 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +19,1,text,stimulated osteochondral discs was significantly higher than that observed in the LF-EMF unstimulate,"[75, 100, 1032, 244]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +19,2,paragraph_title,4. Discussion,"[76, 323, 240, 352]",section_heading,0.85,paragraph_title label with numbering: 4. Discussion,section_heading,0.85,body_zone,heading_like,heading_numbered,True,True +19,3,text,"In this study, we proposed a novel means of cartilage repair that uses MR-SCS combined with EMA and ","[74, 390, 1033, 1465]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +20,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p20 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +20,1,text,"cell characteristics and functions [29]. In contrast, the 3D culture of MSCs possesses many advantag","[74, 101, 1033, 963]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +20,2,text,"Many studies have the reported toxicity issues of MNPs on MSCs, and most of them resulted non- or ve","[74, 985, 1033, 1400]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +21,0,header,Journal Pre-proof,"[504, 1, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p21 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +21,1,text," $ \mu g $, and these concentrations are relatively low compared to those used by the other research","[74, 100, 1028, 186]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +21,2,text,"It is known that MNPs internalize within cells through several endocytosis pathway, such as such as ","[74, 216, 1033, 1457]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +22,0,header,Journal Pre-proof,"[504, 1, 832, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p22 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +22,1,text,"materials, respective affinity for iron species of the soluble chelators and surface ligands of the ","[74, 101, 1033, 524]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +22,2,text,"Additionally, controversy remains regarding the possible inhibitory effect of MNP labeling on chondr","[74, 543, 1032, 1073]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +22,3,text,"Moreover, external biophysical stimulation by LF-EMF was used successfully for the regeneration of a","[75, 1095, 1033, 1456]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +23,0,paragraph_title,Journal Pre-proof,"[504, 1, 831, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p23 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +23,1,text,external biophysical stimulation by LF-EMF was used successfully for the regeneration of articular c,"[73, 97, 1036, 1456]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +24,0,header,Journal Pre-proof,"[504, 1, 831, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p24 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +24,1,text,MR-SCS appear as continuous tissue in microscopic images obtained after LF-EMF stimulation and in un,"[75, 99, 1033, 909]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +24,2,paragraph_title,4. Conclusion,"[76, 986, 247, 1016]",section_heading,0.85,paragraph_title label with numbering: 4. Conclusion,section_heading,0.85,body_zone,heading_like,heading_numbered,True,True +24,3,text,We developed a magnetoresponsive stem cell spheroid-based cartilage recovery platform that can preci,"[75, 1050, 1033, 1411]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,body_zone,body_like,none,True,True +25,0,header,Journal Pre-proof,"[504, 2, 830, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p25 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +25,1,text,fields can provide effective potential therapeutic agents to promote the regeneration of damaged tis,"[75, 100, 1031, 186]",body_paragraph,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +25,2,paragraph_title,Conflict of Interest,"[76, 241, 311, 270]",section_heading,0.5,backmatter heading on page 25: Conflict of Interest,backmatter_heading_candidate,0.8,tail_nonref_hold_zone,support_like,none,True,True +25,3,text,All other authors declare that they have no competing interests.,"[75, 359, 690, 387]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +25,4,paragraph_title,Acknowledgments,"[76, 444, 304, 474]",section_heading,0.5,backmatter heading on page 25: Acknowledgments,backmatter_heading_candidate,0.8,tail_nonref_hold_zone,heading_like,short_fragment,True,True +25,5,text,A. Yoo and G. Go contributed equally to this work. We thank Medical Microrobot Center of Chonnam Nat,"[74, 518, 1032, 881]",backmatter_body,0.6,default body_paragraph for text label; tail_nonref_hold_zone excluded from body flow; tail_nonref_hold_zone excluded from body flow,backmatter_body,0.6,tail_nonref_hold_zone,unknown_like,none,True,True +26,0,header,Journal Pre-proof,"[503, 2, 831, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p26 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +26,1,paragraph_title,References,"[75, 102, 216, 133]",reference_heading,0.9,references heading: References,reference_heading,0.9,,heading_like,short_fragment,True,True +26,2,reference_content,"[1] C. Cooper, M.K. Javaid, N. Arden, Epidemiology of osteoarthritis, Atlas of osteoarthritis, Sprin","[78, 166, 1027, 249]",reference_item,0.85,"reference content label: [1] C. Cooper, M.K. Javaid, N. Arden, Epidemiology of osteoa",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,3,reference_content,"[2] P. Creamer, M. Hochberg, Why does osteoarthritis of the knee hurt--sometimes?, Rheumatology, 36(","[77, 275, 1027, 360]",reference_item,0.85,"reference content label: [2] P. Creamer, M. Hochberg, Why does osteoarthritis of the ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,4,reference_content,"[3] L. Zhang, J. Hu, K.A. Athanasiou, The role of tissue engineering in articular cartilage repair a","[78, 386, 1028, 470]",reference_item,0.85,"reference content label: [3] L. Zhang, J. Hu, K.A. Athanasiou, The role of tissue eng",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,5,reference_content,"[4] I. Uzieliene, P. Bernotas, A. Mobasheri, E. Bernotiene, The Role of Physical Stimuli on Calcium ","[79, 496, 1026, 581]",reference_item,0.85,"reference content label: [4] I. Uzieliene, P. Bernotas, A. Mobasheri, E. Bernotiene, ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,6,reference_content,"[5] E.B. Hunziker, Articular cartilage repair: basic science and clinical progress. A review of the ","[79, 607, 1027, 690]",reference_item,0.85,"reference content label: [5] E.B. Hunziker, Articular cartilage repair: basic science",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,7,reference_content,"[6] C. Baugé, K. Boumédiene, Use of Adult Stem Cells for Cartilage Tissue Engineering: Current Statu","[79, 717, 1027, 801]",reference_item,0.85,"reference content label: [6] C. Baugé, K. Boumédiene, Use of Adult Stem Cells for Car",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,8,reference_content,"[7] A. Khademhosseini, Micro and nanoengineering of the cell microenvironment: technologies and appl","[79, 827, 1026, 911]",reference_item,0.85,"reference content label: [7] A. Khademhosseini, Micro and nanoengineering of the cell",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,9,reference_content,"[8] H. Rubash, J. Berry, Revisions of hip and knee replacements in Canada, Canadian Joint Replacemen","[77, 938, 1027, 1075]",reference_item,0.85,"reference content label: [8] H. Rubash, J. Berry, Revisions of hip and knee replaceme",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,10,reference_content,"[9] Y. Petrenko, E. Syková, Š. Kubinová, The therapeutic potential of three-dimensional multipotent ","[77, 1102, 1027, 1189]",reference_item,0.85,"reference content label: [9] Y. Petrenko, E. Syková, Š. Kubinová, The therapeutic pot",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +26,11,reference_content,"[10] H. Dashtdar, H. Rothan, T. Tay, R.E. Raja Ahmad, R. Ali, L. Tay, et al., A Preliminary Study Co","[79, 1214, 1028, 1404]",reference_item,0.85,"reference content label: [10] H. Dashtdar, H. Rothan, T. Tay, R.E. Raja Ahmad, R. Ali",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,0,header,Journal Pre-proof,"[504, 2, 831, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p27 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +27,1,reference_content,"[11] N. Landázuri, S. Tong, J. Suo, G. Joseph, D. Weiss, D.J. Sutcliffe, et al., Magnetic targeting ","[78, 100, 1028, 240]",reference_item,0.85,"reference content label: [11] N. Landázuri, S. Tong, J. Suo, G. Joseph, D. Weiss, D.J",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,2,reference_content,"[12] G. Go, J. Han, J. Zhen, S. Zheng, A. Yoo, M.-J. Jeon, et al., A Magnetically Actuated Microscaf","[77, 266, 1028, 404]",reference_item,0.85,"reference content label: [12] G. Go, J. Han, J. Zhen, S. Zheng, A. Yoo, M.-J. Jeon, e",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,3,reference_content,"[13] S. Jeon, S. Kim, S. Ha, S. Lee, E. Kim, S.Y. Kim, et al., Magnetically actuated microrobots as ","[79, 432, 1025, 520]",reference_item,0.85,"reference content label: [13] S. Jeon, S. Kim, S. Ha, S. Lee, E. Kim, S.Y. Kim, et al",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,4,reference_content,"[14] M. van der Elst, C.P.A.T. Klein, J.M. de Blieck-Hogervorst, P. Patka, H.J.T.M. Haarman, Bone ti","[78, 541, 1028, 683]",reference_item,0.85,"reference content label: [14] M. van der Elst, C.P.A.T. Klein, J.M. de Blieck-Hogervo",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,5,reference_content,"[15] M. J Martin, A. Muotri, F. Gage, A. Varki, Human embryonic stem cells express an immunogenic no","[76, 708, 1027, 793]",reference_item,0.85,"reference content label: [15] M. J Martin, A. Muotri, F. Gage, A. Varki, Human embryo",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,6,reference_content,"[16] W. Mueller-Klieser, Three-dimensional cell cultures: From molecular mechanisms to clinical appl","[77, 818, 1028, 904]",reference_item,0.85,"reference content label: [16] W. Mueller-Klieser, Three-dimensional cell cultures: Fr",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,7,reference_content,"[17] D. Ciombor, G. Lester, R. Aaron, P. Neame, B. Caterson, Low Frequency EMF Regulates Chondrocyte","[76, 929, 1027, 1014]",reference_item,0.85,"reference content label: [17] D. Ciombor, G. Lester, R. Aaron, P. Neame, B. Caterson,",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,8,reference_content,"[18] S. Mayer-Wagner, A. Passberger, B. Sievers, J. Aigner, B. Summer, T.S. Schiergens, et al., Effe","[77, 1040, 1028, 1180]",reference_item,0.85,"reference content label: [18] S. Mayer-Wagner, A. Passberger, B. Sievers, J. Aigner, ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +27,9,reference_content,"[19] J. Nowak, F. Wiekhorst, L. Trahms, S. Odenbach, The influence of hydrodynamic diameter and core","[76, 1206, 1031, 1345]",reference_item,0.85,"reference content label: [19] J. Nowak, F. Wiekhorst, L. Trahms, S. Odenbach, The inf",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,0,header,Journal Pre-proof,"[503, 2, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p28 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +28,1,reference_content,"[20] M. Kallumadil, M. Tada, T. Nakagawa, M. Abe, P. Southern, Q.A. Pankhurst, Suitability of commer","[79, 100, 1028, 239]",reference_item,0.85,"reference content label: [20] M. Kallumadil, M. Tada, T. Nakagawa, M. Abe, P. Souther",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,2,reference_content,"[21] C. Ross, Mechanisms of extra low frequency electromagnetic field (ELF-EMF) on human bone marrow","[78, 266, 1027, 352]",reference_item,0.85,"reference content label: [21] C. Ross, Mechanisms of extra low frequency electromagne",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,3,reference_content,"[22] J.E. Henrietta, M.-J.M. Eileen, ON THE VISCOSITY AND pH OF SYNOVIAL FLUID AND THE pH OF BLOOD, ","[77, 378, 1027, 516]",reference_item,0.85,"reference content label: [22] J.E. Henrietta, M.-J.M. Eileen, ON THE VISCOSITY AND pH",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,4,reference_content,"[23] D. Choudhury, R. Walker, T. Roy, S. Paul, R. Mootanah, Performance of honed Surface profiles to","[77, 543, 1027, 629]",reference_item,0.85,"reference content label: [23] D. Choudhury, R. Walker, T. Roy, S. Paul, R. Mootanah, ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,5,reference_content,"[24] C.J. Centeno, D. Busse, J. Kisiday, C. Keohan, M. Freeman, Increased knee cartilage volume in d","[77, 654, 1027, 795]",reference_item,0.85,"reference content label: [24] C.J. Centeno, D. Busse, J. Kisiday, C. Keohan, M. Freem",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,6,reference_content,"[25] U. Nöth, A.F. Steinert, R.S. Tuan, Technology Insight: adult mesenchymal stem cells for osteoar","[79, 820, 1027, 904]",reference_item,0.85,"reference content label: [25] U. Nöth, A.F. Steinert, R.S. Tuan, Technology Insight: ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,7,reference_content,[26] http://www.medi-post.com/front/eng/stemcell/cartistem.do.,"[78, 930, 699, 961]",reference_item,0.85,reference content label: [26] http://www.medi-post.com/front/eng/stemcell/cartistem.d,reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,8,reference_content,"[27] G. Kamei, T. Kobayashi, S. Ohkawa, W. Kongcharoensombat, N. Adachi, K. Takazawa, et al., Articu","[79, 985, 1027, 1126]",reference_item,0.85,"reference content label: [27] G. Kamei, T. Kobayashi, S. Ohkawa, W. Kongcharoensombat",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,9,reference_content,"[28] E.E. Mahmoud, G. Kamei, Y. Harada, R. Shimizu, N. Kamei, N. Adachi, et al., Cell magnetic targe","[77, 1150, 1027, 1291]",reference_item,0.85,"reference content label: [28] E.E. Mahmoud, G. Kamei, Y. Harada, R. Shimizu, N. Kamei",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +28,10,reference_content,"[29] P.R. Baraniak, T.C. McDevitt, Scaffold-free culture of mesenchymal stem cell spheroids in suspe","[78, 1317, 1028, 1401]",reference_item,0.85,"reference content label: [29] P.R. Baraniak, T.C. McDevitt, Scaffold-free culture of ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,0,header,Journal Pre-proof,"[503, 2, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p29 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +29,1,reference_content,"[30] Y. Yamaguchi, J. Ohno, A. Sato, H. Kido, T. Fukushima, Mesenchymal stem cell spheroids exhibit ","[78, 100, 1028, 239]",reference_item,0.85,"reference content label: [30] Y. Yamaguchi, J. Ohno, A. Sato, H. Kido, T. Fukushima, ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,2,reference_content,"[31] U. De Simone, M. Roccio, L. Gribaldo, A. Spinillo, F. Caloni, T. Coccini, Human 3D Cultures as ","[77, 266, 1028, 405]",reference_item,0.85,"reference content label: [31] U. De Simone, M. Roccio, L. Gribaldo, A. Spinillo, F. C",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,3,reference_content,"[32] J. Riegler, A. Liew, S.O. Hynes, D. Ortega, T. O'Brien, R.M. Day, et al., Superparamagnetic iro","[79, 431, 1026, 517]",reference_item,0.85,"reference content label: [32] J. Riegler, A. Liew, S.O. Hynes, D. Ortega, T. O'Brien,",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,4,reference_content,"[33] S. Saha, X. Yang, S. Tanner, S. Curran, D. Wood, J. Kirkham, The effects of iron oxide incorpor","[79, 541, 1027, 627]",reference_item,0.85,"reference content label: [33] S. Saha, X. Yang, S. Tanner, S. Curran, D. Wood, J. Kir",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,5,reference_content,"[34] M. Mahmoudi, H. Hofmann, B. Rothen-Rutishauser, A. Petri-Fink, Assessing the in vitro and in vi","[77, 652, 1027, 791]",reference_item,0.85,"reference content label: [34] M. Mahmoudi, H. Hofmann, B. Rothen-Rutishauser, A. Petr",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,6,reference_content,"[35] C. Wilhelm, F. Gazeau, Universal cell labelling with anionic magnetic nanoparticles, Biomateria","[76, 818, 1027, 903]",reference_item,0.85,"reference content label: [35] C. Wilhelm, F. Gazeau, Universal cell labelling with an",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,7,reference_content,"[36] S. Salatin, A. Yari Khosroushahi, Overviews on the cellular uptake mechanism of polysaccharide ","[76, 929, 1028, 1067]",reference_item,0.85,"reference content label: [36] S. Salatin, A. Yari Khosroushahi, Overviews on the cell",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,8,reference_content,"[37] K. Andreas, R. Georgieva, M. Ladwig, S. Mueller, M. Notter, M. Sittinger, et al., Highly effici","[77, 1094, 1028, 1235]",reference_item,0.85,"reference content label: [37] K. Andreas, R. Georgieva, M. Ladwig, S. Mueller, M. Not",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,9,reference_content,"[38] G.n. Jutz, P. van Rijn, B. Santos Miranda, A. Böker, Ferritin: a versatile building block for b","[79, 1259, 1027, 1345]",reference_item,0.85,"reference content label: [38] G.n. Jutz, P. van Rijn, B. Santos Miranda, A. Böker, Fe",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +29,10,reference_content,"[39] S. Sharifi, S. Behzadi, S. Laurent, M.L. Forrest, P. Stroeve, M. Mahmoudi, Toxicity of nanomate","[79, 1370, 1028, 1452]",reference_item,0.85,"reference content label: [39] S. Sharifi, S. Behzadi, S. Laurent, M.L. Forrest, P. St",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,0,header,Journal Pre-proof,"[504, 2, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p30 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +30,1,reference_content,"[40] P. Arosio, R. Ingrassia, P. Cavadini, Ferritins: a family of molecules for iron storage, antiox","[78, 101, 1028, 238]",reference_item,0.85,"reference content label: [40] P. Arosio, R. Ingrassia, P. Cavadini, Ferritins: a fami",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,2,reference_content,"[41] G. Zhao, F. Bou-Abdallah, P. Arosio, S. Levi, C. Janus-Chandler, N.D. Chasteen, Multiple pathwa","[77, 266, 1027, 404]",reference_item,0.85,"reference content label: [41] G. Zhao, F. Bou-Abdallah, P. Arosio, S. Levi, C. Janus-",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,3,reference_content,"[42] A.S. Arbab, L.B. Wilson, P. Ashari, E.K. Jordan, B.K. Lewis, J.A. Frank, A model of lysosomal m","[77, 432, 1028, 573]",reference_item,0.85,"reference content label: [42] A.S. Arbab, L.B. Wilson, P. Ashari, E.K. Jordan, B.K. L",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,4,reference_content,"[43] M. Lévy, F. Lagarde, V.-A. Maraloiu, M.-G. Blanchin, F. Gendron, C. Wilhelm, et al., Degradabil","[77, 597, 1027, 739]",reference_item,0.85,"reference content label: [43] M. Lévy, F. Lagarde, V.-A. Maraloiu, M.-G. Blanchin, F.",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,5,reference_content,"[44] T. Skotland, P.C. Sontum, I. Oulie, In vitro stability analyses as a model for metabolism of fe","[77, 763, 1029, 904]",reference_item,0.85,"reference content label: [44] T. Skotland, P.C. Sontum, I. Oulie, In vitro stability ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,6,reference_content,"[45] F. Mazuel, A. Espinosa, N. Luciani, M. Reffay, R. Le Borgne, L. Motte, et al., Massive intracel","[76, 930, 1028, 1068]",reference_item,0.85,"reference content label: [45] F. Mazuel, A. Espinosa, N. Luciani, M. Reffay, R. Le Bo",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,7,reference_content,"[46] Y.-K. Chang, Y.-P. Liu, J. H Ho, S.-C. Hsu, O. Lee, Amine-surface-modified superparamagnetic ir","[77, 1094, 1028, 1234]",reference_item,0.85,"reference content label: [46] Y.-K. Chang, Y.-P. Liu, J. H Ho, S.-C. Hsu, O. Lee, Ami",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +30,8,reference_content,"[47] A. Heymer, D. Haddad, M. Weber, U. Gbureck, P.M. Jakob, J. Eulert, et al., Iron oxide labelling","[77, 1261, 1027, 1397]",reference_item,0.85,"reference content label: [47] A. Heymer, D. Haddad, M. Weber, U. Gbureck, P.M. Jakob,",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,0,header,Journal Pre-proof,"[503, 2, 831, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p31 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +31,1,reference_content,"[48] X.-h. Jing, L. Yang, X.-j. Duan, B. Xie, W. Chen, Z. Li, et al., In vivo MRI tracking of magnet","[78, 100, 1028, 240]",reference_item,0.85,"reference content label: [48] X.-h. Jing, L. Yang, X.-j. Duan, B. Xie, W. Chen, Z. Li",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,2,reference_content,"[49] L.-Y. Sun, D.-K. Hsieh, P.-C. Lin, H.-T. Chiu, T.-W. Chiou, Pulsed electromagnetic fields accel","[77, 266, 1028, 405]",reference_item,0.85,"reference content label: [49] L.-Y. Sun, D.-K. Hsieh, P.-C. Lin, H.-T. Chiu, T.-W. Ch",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,3,reference_content,"[50] S. Mayer-Wagner, A. Passberger, B. Sievers, J. Aigner, B. Summer, T. Schiergens, et al., Effect","[77, 432, 1028, 572]",reference_item,0.85,"reference content label: [50] S. Mayer-Wagner, A. Passberger, B. Sievers, J. Aigner, ",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,4,reference_content,"[51] A. Maziarz, B. Kocan, M. Bester, S. Budzik, M. Cholewa, T. Ochiya, et al., How electromagnetic ","[79, 597, 1027, 681]",reference_item,0.85,"reference content label: [51] A. Maziarz, B. Kocan, M. Bester, S. Budzik, M. Cholewa,",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,5,reference_content,"[52] H. Akiyama, Control of chondrogenesis by the transcription factor SOX92008.","[78, 707, 887, 737]",reference_item,0.85,"reference content label: [52] H. Akiyama, Control of chondrogenesis by the transcript",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,6,reference_content,"[53] C. Vallbona, C.F. Hazlewood, G. Jurida, Response of pain to static magnetic fields in postpolio","[79, 763, 1027, 901]",reference_item,0.85,"reference content label: [53] C. Vallbona, C.F. Hazlewood, G. Jurida, Response of pai",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,7,reference_content,"[54] H. Amin, M. Brady, J.-P. St-Pierre, M. M. Stevens, D. R. Overby, C. Ethier, Stimulation of Chon","[75, 929, 1030, 1068]",reference_item,0.85,"reference content label: [54] H. Amin, M. Brady, J.-P. St-Pierre, M. M. Stevens, D. R",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,8,reference_content,"[55] C.-H. Chen, Y.-S. Lin, Y.-C. Fu, C.K. Wang, S.-C. Wu, G.-J. Wang, et al., Electromagnetic field","[76, 1093, 1028, 1232]",reference_item,0.85,"reference content label: [55] C.-H. Chen, Y.-S. Lin, Y.-C. Fu, C.K. Wang, S.-C. Wu, G",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,9,reference_content,"[56] S. Muramatsu, M. Wakabayashi, T. Ohno, K. Amano, R. Ooishi, T. Sugahara, et al., Functional Gen","[79, 1260, 1027, 1344]",reference_item,0.85,"reference content label: [56] S. Muramatsu, M. Wakabayashi, T. Ohno, K. Amano, R. Ooi",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +31,10,reference_content,"[57] F. Torossian, A. Bisson, J.-P. Vannier, O. Boyer, M. Lamacz, TRPC expression in mesenchymal ste","[79, 1369, 1028, 1452]",reference_item,0.85,"reference content label: [57] F. Torossian, A. Bisson, J.-P. Vannier, O. Boyer, M. La",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +32,0,header,Journal Pre-proof,"[503, 2, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p32 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +32,1,reference_content,"[58] R.G. LeBaron, K.A. Athanasiou, Ex vivo synthesis of articular cartilage, Biomaterials, 21(2000)","[77, 99, 1028, 183]",reference_item,0.85,"reference content label: [58] R.G. LeBaron, K.A. Athanasiou, Ex vivo synthesis of art",reference_item,0.85,reference_zone,reference_like,reference_numeric_bracket,True,True +32,2,paragraph_title,Biographies,"[81, 350, 234, 380]",sub_subsection_heading,0.6,"unnumbered paragraph_title, inferred level sub_subsection_heading: Biographies",sub_subsection_heading,0.6,,heading_like,short_fragment,True,True +32,3,text,Amr Yoo received her B.S. (2010) degrees from the Dep. of Applied Bioscience and Biotechnology at Ch,"[75, 382, 1025, 511]",body_paragraph,0.6,default body_paragraph for text label; promoted in tail spread from body_paragraph,body_paragraph,0.6,reference_zone,unknown_like,none,True,True +32,4,reference_content,Gwangjun Go received his B.S. (2013) and M.S. (2015) degree from the School of Mechanical Engineerin,"[75, 513, 1018, 619]",reference_item,0.85,reference content label: Gwangjun Go received his B.S. (2013) and M.S. (2015) degree ,reference_item,0.85,reference_zone,unknown_like,none,True,True +32,5,reference_content,Kim Tien Nguyen received a B.S. degree in Dept. of Mechanical Engineering from the Hochiminh Univers,"[75, 620, 1021, 777]",reference_item,0.85,reference content label: Kim Tien Nguyen received a B.S. degree in Dept. of Mechanica,reference_item,0.85,,unknown_like,none,True,True +32,6,reference_content,Kyungmin Lee received her B.S. (2013) and M.S. (2018) degrees from the Dept. of Mechatronics Enginee,"[75, 779, 1021, 937]",reference_item,0.85,reference content label: Kyungmin Lee received her B.S. (2013) and M.S. (2018) degree,reference_item,0.85,reference_zone,unknown_like,none,True,True +32,7,reference_content,Hyun-Ki Min received his B.S. (2012) degrees from the Division of Biological Science at Wonkwang Uni,"[75, 937, 994, 1018]",reference_item,0.85,reference content label: Hyun-Ki Min received his B.S. (2012) degrees from the Divisi,reference_item,0.85,,unknown_like,none,True,True +33,0,header,Journal Pre-proof,"[503, 2, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p33 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +33,1,reference_content,of Medical Microrobotics (KIMIRo). His research interests are fuse micro/nanorobot with biology and ,"[74, 101, 1027, 153]",reference_item,0.85,reference content label: of Medical Microrobotics (KIMIRo). His research interests ar,reference_item,0.85,,unknown_like,none,True,True +33,2,reference_content,Byungjeon Kang received his B.S. (2008) and M.S. (2010) degrees in mechanical engineering from the C,"[76, 155, 1014, 284]",reference_item,0.85,reference content label: Byungjeon Kang received his B.S. (2008) and M.S. (2010) degr,reference_item,0.85,reference_zone,unknown_like,none,True,True +33,3,reference_content,"Chang-Sei Kim received his B.S. (1998), M.S. (2000), and Ph.D. (2011) degrees from the Dept. of Cont","[75, 286, 1019, 472]",reference_item,0.85,"reference content label: Chang-Sei Kim received his B.S. (1998), M.S. (2000), and Ph.",reference_item,0.85,reference_zone,unknown_like,none,True,True +33,4,reference_content,"Jiwon Han received his B.S. (1998), M.S. (2008), and Ph.D. (2015) degrees from the Dep. of Veterinar","[75, 473, 1025, 603]",reference_item,0.85,"reference content label: Jiwon Han received his B.S. (1998), M.S. (2008), and Ph.D. (",reference_item,0.85,reference_zone,unknown_like,none,True,True +33,5,reference_content,Jong-Oh Park received his B.S. (1978) and M.S. (1981) degrees from the department of mechanical engi,"[74, 604, 1029, 791]",reference_item,0.85,reference content label: Jong-Oh Park received his B.S. (1978) and M.S. (1981) degree,reference_item,0.85,reference_zone,unknown_like,none,True,True +34,0,header,Journal Pre-proof,"[504, 2, 831, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p34 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +34,1,text,Engineering and a president of Korea Institute of Medical Microrobotics (KIMIRo). His research inter,"[82, 100, 1032, 449]",body_paragraph,0.6,default body_paragraph for text label,body_paragraph,0.6,reference_zone,unknown_like,none,True,True +35,0,header,Journal Pre-proof,"[505, 1, 830, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p35 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +35,1,paragraph_title,Table and Figure Captions,"[76, 102, 406, 135]",subsection_heading,0.6,"unnumbered paragraph_title, inferred level subsection_heading: Table and Figure Captions",subsection_heading,0.6,,table_caption_like,none,True,True +35,2,figure_title,Fig. 1. Principle idea of magnetoresponsive stem cell spheroid (MR-SCS)-based cartilage recovery pla,"[74, 177, 1028, 373]",figure_caption_candidate,0.92,figure_title label: Fig. 1. Principle idea of magnetoresponsive stem cell sphero,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +35,3,figure_title,Fig. 2. Characterization and chondrogenic differentiation of MR-SCS. (a) Microscopy images and avera,"[74, 413, 1030, 942]",figure_caption_candidate,0.92,figure_title label: Fig. 2. Characterization and chondrogenic differentiation of,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +35,4,figure_title,Fig. 3. Magnetic actuation characterization of MR-SCS. (a) Experimental setup of the EMA system. (b),"[75, 979, 1032, 1422]",figure_caption_candidate,0.92,figure_title label: Fig. 3. Magnetic actuation characterization of MR-SCS. (a) E,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +36,0,header,Journal Pre-proof,"[503, 1, 832, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p36 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +36,1,figure_title,Fig. 4. Effect of LF-EMF stimulation on MR-SCS. (a) Experimental setup of the LF-EMF stimulation sys,"[74, 99, 1032, 421]",figure_caption_candidate,0.92,figure_title label: Fig. 4. Effect of LF-EMF stimulation on MR-SCS. (a) Experime,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +37,0,header,Journal Pre-proof,"[504, 1, 830, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p37 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +37,1,figure_title,(a),"[118, 107, 148, 133]",figure_caption_candidate,0.85,figure_title label: (a),figure_caption,0.85,,unknown_like,short_fragment,False,False +37,2,image,,"[114, 99, 1025, 410]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +37,3,figure_title,(c),"[707, 106, 739, 135]",figure_caption_candidate,0.85,figure_title label: (c),figure_caption,0.85,,unknown_like,short_fragment,False,False +37,4,figure_title,Fig. 1. Principle idea of magnetoresponsive stem cell spheroid (MR-SCS)-based cartilage recovery pla,"[74, 478, 1032, 679]",figure_caption_candidate,0.92,figure_title label: Fig. 1. Principle idea of magnetoresponsive stem cell sphero,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +38,0,header,Journal Pre-proof,"[505, 2, 830, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p38 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +38,1,figure_title,(a),"[94, 109, 128, 138]",figure_caption_candidate,0.85,figure_title label: (a),figure_caption,0.85,,unknown_like,short_fragment,False,False +38,2,image,,"[91, 146, 183, 211]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,3,figure_title,0 mg/mL,"[110, 210, 165, 225]",figure_caption_candidate,0.85,figure_title label: 0 mg/mL,figure_caption,0.85,reference_zone,reference_like,reference_numeric_dot,False,False +38,4,image,,"[92, 232, 182, 304]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,5,image,,"[192, 144, 283, 209]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,6,figure_title,0.3 mg/mL,"[107, 298, 169, 312]",figure_caption_candidate,0.85,figure_title label: 0.3 mg/mL,figure_caption,0.85,,unknown_like,short_fragment,False,False +38,7,figure_title,0.1 mg/mL,"[209, 210, 274, 226]",figure_caption_candidate,0.85,figure_title label: 0.1 mg/mL,figure_caption,0.85,,unknown_like,short_fragment,False,False +38,8,image,,"[193, 231, 284, 299]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,9,image,,"[291, 140, 383, 220]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,10,figure_title,0.4 mg/mL,"[209, 298, 271, 312]",figure_caption_candidate,0.85,figure_title label: 0.4 mg/mL,figure_caption,0.85,,unknown_like,short_fragment,False,False +38,11,image,,"[295, 227, 386, 301]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,12,chart,,"[399, 118, 636, 327]",media_asset,0.85,media label: chart,media_asset,0.85,,unknown_like,empty,True,True +38,13,figure_title,(b),"[620, 109, 655, 137]",figure_caption_candidate,0.85,figure_title label: (b),figure_caption,0.85,,unknown_like,short_fragment,False,False +38,14,chart,,"[628, 106, 990, 317]",media_asset,0.85,media label: chart,media_asset,0.85,,unknown_like,empty,True,True +38,15,figure_title,(c) Nuclei / Cytoplasm / Iron (ferric form),"[93, 339, 404, 363]",figure_caption_candidate,0.85,figure_title label: (c) Nuclei / Cytoplasm / Iron (ferric form),figure_caption,0.85,,legend_like,none,False,False +38,16,figure_title,(d),"[427, 336, 462, 364]",figure_caption_candidate,0.85,figure_title label: (d),figure_caption,0.85,,unknown_like,short_fragment,False,False +38,17,image,,"[93, 341, 413, 616]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,18,image,,"[427, 337, 577, 620]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,19,image,,"[590, 341, 700, 621]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,20,image,,"[701, 341, 809, 622]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,21,image,,"[811, 341, 918, 621]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,22,image,,"[923, 337, 1028, 620]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,23,figure_title,(e),"[95, 637, 127, 667]",figure_caption_candidate,0.85,figure_title label: (e),figure_caption,0.85,,unknown_like,short_fragment,False,False +38,24,image,,"[101, 635, 236, 911]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,25,figure_title,(f),"[243, 636, 270, 667]",figure_caption_candidate,0.85,figure_title label: (f),figure_caption,0.85,,unknown_like,short_fragment,False,False +38,26,image,,"[248, 640, 393, 912]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,27,image,,"[396, 646, 519, 911]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,28,figure_title,(g),"[648, 637, 683, 667]",figure_caption_candidate,0.85,figure_title label: (g),figure_caption,0.85,,unknown_like,short_fragment,False,False +38,29,image,,"[522, 648, 641, 912]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +38,30,chart,,"[652, 668, 1018, 919]",media_asset,0.85,media label: chart,media_asset,0.85,,unknown_like,empty,True,True +38,31,figure_title,Fig. 2. Characterization and chondrogenic differentiation of MR-SCS. (a) Microscopy images and avera,"[73, 976, 1032, 1503]",figure_caption_candidate,0.92,figure_title label: Fig. 2. Characterization and chondrogenic differentiation of,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +39,0,header,Journal Pre-proof,"[504, 2, 829, 39]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p39 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +39,1,figure_title,(a),"[264, 110, 291, 135]",figure_caption_candidate,0.85,figure_title label: (a),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,2,image,,"[282, 138, 590, 374]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,3,figure_title,(b),"[593, 110, 620, 134]",figure_caption_candidate,0.85,figure_title label: (b),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,4,chart,,"[595, 142, 856, 383]",media_asset,0.85,media label: chart,media_asset,0.85,,unknown_like,empty,True,True +39,5,figure_title,(c),"[265, 389, 291, 413]",figure_caption_candidate,0.85,figure_title label: (c),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,6,image,,"[297, 411, 516, 611]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,7,figure_title,(d),"[521, 388, 548, 412]",figure_caption_candidate,0.85,figure_title label: (d),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,8,image,,"[296, 623, 517, 799]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,9,image,,"[537, 411, 859, 665]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,10,image,,"[536, 670, 855, 801]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,11,figure_title,(e),"[264, 813, 291, 838]",figure_caption_candidate,0.85,figure_title label: (e),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,12,figure_title,(f),"[526, 814, 548, 838]",figure_caption_candidate,0.85,figure_title label: (f),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,13,image,,"[264, 817, 535, 1150]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,14,image,,"[554, 816, 855, 1143]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,15,figure_title,(g),"[264, 1152, 291, 1178]",figure_caption_candidate,0.85,figure_title label: (g),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,16,figure_title,(h),"[550, 1152, 578, 1177]",figure_caption_candidate,0.85,figure_title label: (h),figure_caption,0.85,,unknown_like,short_fragment,False,False +39,17,image,,"[225, 1153, 556, 1443]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +39,18,figure_title,Knee cartilage,"[351, 1424, 447, 1443]",figure_caption_candidate,0.85,figure_title label: Knee cartilage,figure_caption,0.85,,unknown_like,short_fragment,False,False +39,19,image,,"[573, 1190, 850, 1428]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +40,0,header,Journal Pre-proof,"[503, 1, 831, 41]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p40 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +40,1,figure_title,Fig. 3. Magnetic actuation characterization of MR-SCS. (a) Experimental setup of the EMA system. (b),"[73, 99, 1032, 548]",figure_caption_candidate,0.92,figure_title label: Fig. 3. Magnetic actuation characterization of MR-SCS. (a) E,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +41,0,header,Journal Pre-proof,"[504, 2, 829, 40]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p41 y=2/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False +41,1,figure_title,(a),"[234, 107, 266, 136]",figure_caption_candidate,0.85,figure_title label: (a),figure_caption,0.85,,unknown_like,short_fragment,False,False +41,2,image,,"[236, 110, 868, 381]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +41,3,figure_title,(b),"[232, 398, 265, 426]",figure_caption_candidate,0.85,figure_title label: (b),figure_caption,0.85,,unknown_like,short_fragment,False,False +41,4,image,,"[255, 400, 576, 618]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +41,5,figure_title,(c),"[576, 400, 606, 426]",figure_caption_candidate,0.85,figure_title label: (c),figure_caption,0.85,,unknown_like,short_fragment,False,False +41,6,chart,,"[589, 402, 859, 639]",media_asset,0.85,media label: chart,media_asset,0.85,,unknown_like,empty,True,True +41,7,figure_title,(d),"[233, 642, 266, 671]",figure_caption_candidate,0.85,figure_title label: (d),figure_caption,0.85,,unknown_like,short_fragment,False,False +41,8,image,,"[258, 650, 551, 873]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +41,9,image,,"[567, 658, 854, 873]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +41,10,figure_title,(e),"[232, 889, 261, 918]",figure_caption_candidate,0.85,figure_title label: (e),figure_caption,0.85,,unknown_like,short_fragment,False,False +41,11,chart,,"[241, 889, 566, 1112]",media_asset,0.85,media label: chart,media_asset,0.85,,unknown_like,empty,True,True +41,12,image,,"[601, 882, 851, 1107]",media_asset,0.85,media label: image,media_asset,0.85,,unknown_like,empty,True,True +41,13,figure_title,Fig. 4. Effect of LF-EMF stimulation on MR-SCS. (a) Experimental setup of the LF-EMF stimulation sys,"[74, 1169, 1032, 1491]",figure_caption_candidate,0.92,figure_title label: Fig. 4. Effect of LF-EMF stimulation on MR-SCS. (a) Experime,figure_caption,0.92,display_zone,legend_like,figure_number,False,False +42,0,header,Journal Pre-proof,"[502, 1, 832, 42]",frontmatter_noise,0.98,journal pre-proof running header suppressed: p42 y=1/1584,frontmatter_noise,0.98,preproof_cover_zone,unknown_like,preproof_marker,False,False diff --git a/tests/fixtures/ocr_real_papers/DWQQK2YB/expectations.json b/tests/fixtures/ocr_real_papers/DWQQK2YB/expectations.json index b4b8e04f..ee7c9315 100644 --- a/tests/fixtures/ocr_real_papers/DWQQK2YB/expectations.json +++ b/tests/fixtures/ocr_real_papers/DWQQK2YB/expectations.json @@ -1,22 +1,243 @@ { "document": { - "expected_reader_figure_count_min": 3, + "expected_reader_figure_count_min": 4, "expected_reference_zone": true, - "expected_abstract_span": true + "expected_abstract_span": true, + "notes": [ + "IDEAL expectations for DWQQK2YB (Magnetoresponsive Stem Cell Spheroid, Elsevier 2019).", + "raw_label can be paragraph_title; final role should be section_heading/subsection_heading.", + "Title/authors belong to frontmatter artifact zones, not body_zone.", + "Annotated page images (i folder) are the visual ground truth." + ] }, "pages": { + "1": { + "assertions": [ + {"text_contains": "Journal Pre-proof", "expected_role": "frontmatter_noise", "expected_zone": "preproof_cover_zone", "notes": "Already correct"}, + {"text_contains": "Magnetoresponsive Stem Cell Spheroid", "expected_role": "paper_title", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: frontmatter_noise - should be paper_title"}, + {"text_contains": "Ami Yoo, Gwangjun Go", "expected_role": "authors", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: frontmatter_noise - should be authors"}, + {"text_contains": "PII:", "expected_role": "frontmatter_support", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true}, + {"text_contains": "To appear in:", "expected_role": "frontmatter_support", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true}, + {"text_contains": "Received Date:", "expected_role": "frontmatter_support", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true}, + {"text_contains": "2019 Published by Elsevier", "expected_role": "frontmatter_support", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true} + ] + }, + "2": { + "assertions": [ + {"text_contains": "Magnetoresponsive Stem Cell Spheroid", "expected_role": "paper_title", "expected_zone": "frontmatter_main_zone", "must_not_render_as_body": true, "notes": "Real: unknown_structural - should be paper_title"}, + {"text_contains": "These authors contributed equally", "expected_role": "structured_insert", "expected_zone_any_of": ["frontmatter_side_zone", "frontmatter_main_zone"], "notes": "Real: frontmatter_noise - should be structured_insert"}, + {"text_contains": "Corresponding author", "expected_role": "structured_insert", "expected_zone_any_of": ["frontmatter_side_zone", "frontmatter_main_zone"], "notes": "Real: frontmatter_noise - should be structured_insert"}, + {"text_equals": "Highlights", "expected_role": "paragraph_title", "expected_zone": "body_zone", "notes": "Real: structured_insert - should be paragraph_title (it's a section label, not a structured insert)"} + ] + }, + "3": { + "assertions": [ + {"text_contains": "Histological evaluation revealed", "expected_role": "body_paragraph", "expected_zone": "body_zone", "notes": "Real: abstract_body - this is Highlights continuation, should be body_paragraph"}, + {"text_equals": "Abstract", "expected_role": "paragraph_title", "expected_zone": "body_zone", "notes": "Real: abstract_heading - downstream consumers need paragraph_title"}, + {"text_contains": "Mesenchymal stem cells", "expected_role": "abstract_body", "expected_zone": "body_zone"}, + {"text_contains": "Keywords:", "expected_role": "structured_insert", "expected_zone": "body_zone", "notes": "Real: body_paragraph - should be structured_insert"}, + {"text_equals": "1. Introduction", "expected_role": "section_heading", "expected_zone": "body_zone", "notes": "Real: section_heading - this is correct! raw_label=paragraph_title, final role=section_heading"} + ] + }, + "4": { + "assertions": [ + {"text_contains": "As the world population increases", "expected_role": "body_paragraph", "expected_zone": "body_zone"}, + {"count_text": {"min": 2, "max": 5}} + ] + }, + "5": { + "assertions": [ + {"text_contains": "tissues and organs", "expected_role": "body_paragraph", "expected_zone": "body_zone"}, + {"count_text": {"min": 2, "max": 5}} + ] + }, + "6": { + "assertions": [ + {"text_contains": "still some unsolved issues", "expected_role": "body_paragraph", "expected_zone": "body_zone"}, + {"count_text": {"min": 2, "max": 5}} + ] + }, + "7": { + "assertions": [ + {"text_equals": "2. Experimental section", "expected_role": "section_heading", "expected_zone": "body_zone", "notes": "Real: section_heading - correct!"}, + {"text_equals": "2.1 Cell culture", "expected_role": "subsection_heading", "expected_zone": "body_zone", "notes": "Real: subsection_heading - correct!"}, + {"text_equals": "2.2 Fabrication and characteristics of MNP labeled MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "8": { + "assertions": [ + {"text_equals": "2.3 Chondrogenic differentiation of the MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"}, + {"text_equals": "2.4 Experimental setup for electromagnetic actuation of the MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "9": { + "assertions": [ + {"text_contains": "The electromagnetic actuation of the MR-SCS", "expected_role": "body_paragraph", "expected_zone": "body_zone"}, + {"text_equals": "2.5 Experimental setup for low-field electromagnetic stimulation of the MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "10": { + "assertions": [ + {"text_equals": "2.6 Ex vivo cartilage defect model and implantation of the MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"}, + {"text_equals": "2.7 Gene expression analysis", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "11": { + "assertions": [ + {"text_equals": "2.8 Histological and immunohistochemical staining analysis", "expected_role": "subsection_heading", "expected_zone": "body_zone"}, + {"text_equals": "2.9 Statistical data analysis", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "12": { + "assertions": [ + {"text_equals": "3. Results", "expected_role": "section_heading", "expected_zone": "body_zone"}, + {"text_equals": "3.1 Morphology and cell viability of the MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "17": { + "assertions": [ + {"text_equals": "3.4 Effects of LF-EMF stimulation on in vitro chondrogenic differentiation of the MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "18": { + "assertions": [ + {"text_equals": "3.5 Effects of LF-EMF stimulation on ex vivo cartilage regeneration of MR-SCS", "expected_role": "subsection_heading", "expected_zone": "body_zone"} + ] + }, + "19": { + "assertions": [ + {"text_equals": "4. Discussion", "expected_role": "section_heading", "expected_zone": "body_zone"} + ] + }, + "24": { + "assertions": [ + {"text_equals": "4. Conclusion", "expected_role": "section_heading", "expected_zone": "body_zone"} + ] + }, + "25": { + "assertions": [ + {"text_equals": "Conflict of Interest", "expected_role": "backmatter_heading", "expected_zone": "tail_nonref_hold_zone", "notes": "Real: section_heading - should be backmatter_heading"}, + {"text_contains": "All other authors declare", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"}, + {"text_equals": "Acknowledgments", "expected_role": "backmatter_heading", "expected_zone": "tail_nonref_hold_zone", "notes": "Real: section_heading - should be backmatter_heading"}, + {"text_contains": "A. Yoo and G. Go contributed equally", "expected_role": "structured_insert", "expected_zone": "tail_nonref_hold_zone", "notes": "Real: backmatter_body - should be structured_insert (equal contribution)"} + ] + }, + "26": { + "assertions": [ + {"text_equals": "References", "expected_role": "reference_heading", "expected_zone": "reference_zone"}, + {"count_reference_content": {"min": 8, "max": 12}} + ] + }, + "27": { + "assertions": [ + {"count_reference_content": {"min": 7, "max": 12}} + ] + }, + "28": { + "assertions": [ + {"count_reference_content": {"min": 7, "max": 12}} + ] + }, + "29": { + "assertions": [ + {"count_reference_content": {"min": 7, "max": 12}} + ] + }, + "30": { + "assertions": [ + {"count_reference_content": {"min": 7, "max": 12}} + ] + }, + "31": { + "assertions": [ + {"count_reference_content": {"min": 7, "max": 12}} + ] + }, + "33": { + "assertions": [ + {"text_equals": "Biographies", "expected_role": "backmatter_heading", "expected_zone": "tail_nonref_hold_zone", "notes": "CRITICAL BUG: Real pipeline absorbs biographies into reference_zone on page 32. Should be in tail_nonref_hold_zone."}, + {"text_contains": "Ami Yoo received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone", "notes": "CRITICAL BUG: Real has role=body_paragraph in reference_zone."}, + {"text_contains": "Gwangjun Go received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone", "notes": "CRITICAL BUG: Real has role=reference_item."}, + {"text_contains": "Kim Tien Nguyen received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"}, + {"text_contains": "Kyungmin Lee received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone", "notes": "CRITICAL BUG: Real has role=reference_item."}, + {"text_contains": "Hyun-Ki Min received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"} + ] + }, + "34": { + "assertions": [ + {"text_contains": "Byungjeon Kang received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"}, + {"text_contains": "Chang-Sei Kim received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"}, + {"text_contains": "Jiwon Han received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"}, + {"text_contains": "Jong-Oh Park received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"}, + {"text_contains": "Eunpyo Choi received", "expected_role": "backmatter_body", "expected_zone": "tail_nonref_hold_zone"} + ] + }, "35": { - "expected_object_ownership": [ - { - "object_type": "figure", - "figure_number": 1, - "must_render_as_object": true, - "must_not_render_caption_blocks_as_body": true - } + "assertions": [ + {"text_equals": "Table and Figure Captions", "expected_role": "backmatter_heading", "expected_zone": "tail_nonref_hold_zone", "notes": "Real: subsection_heading - should be backmatter_heading"}, + {"text_contains": "Fig. 1. Principle idea", "expected_role": "figure_caption", "expected_zone": "display_zone", "notes": "CRITICAL BUG: Real has role=unknown_structural."}, + {"text_contains": "Fig. 2. Characterization", "expected_role": "figure_caption", "expected_zone": "display_zone"}, + {"text_contains": "Fig. 3. Magnetic actuation", "expected_role": "figure_caption", "expected_zone": "display_zone"} + ] + }, + "36": { + "assertions": [ + {"text_contains": "Fig. 4. Effect of LF-EMF", "expected_role": "figure_caption", "expected_zone": "display_zone"} + ] + }, + "37": { + "assertions": [ + {"text_contains": "Fig. 1. Principle idea", "expected_role": "figure_caption", "expected_zone": "display_zone"} + ] + }, + "38": { + "assertions": [ + {"text_contains": "Fig. 2. Characterization", "expected_role": "figure_caption", "expected_zone": "display_zone"}, + {"count_image": {"min": 10, "max": 20}} + ] + }, + "40": { + "assertions": [ + {"text_contains": "Fig. 3. Magnetic actuation", "expected_role": "figure_caption", "expected_zone": "display_zone"} ] } }, - "expected_render_invariants": [ - {"type": "no_duplicate_caption", "caption_regex": "Fig\\.\\s*1"} + "expected_bugs": [ + { + "bug": "biography_absorbed_into_references", + "pages": [33, 34], + "description": "All 10 biographies are absorbed into reference_zone. Ami Yoo has role=body_paragraph, Gwangjun Go/Kyungmin Lee have role=reference_item. Should be backmatter_body in tail_nonref_hold_zone.", + "fix": "Biographies section after last reference should trigger tail spread into tail_nonref_hold_zone" + }, + { + "bug": "figure_captions_not_recognized", + "pages": [35, 36, 37, 38, 40], + "description": "All figure captions (Fig. 1-4) have role=unknown_structural instead of figure_caption.", + "fix": "Figure captions should be figure_caption in display_zone" + }, + { + "bug": "frontmatter_metadata_not_classified", + "pages": [1], + "description": "PII, DOI, To appear in, Received Date, Published by Elsevier are unclassified (no explicit role). Should be frontmatter_support.", + "fix": "Frontmatter support lines need dedicated classification" + }, + { + "bug": "paper_title_not_classified", + "pages": [1, 2], + "description": "Title text gets role=frontmatter_noise (page 1) or unknown_structural (page 2) instead of paper_title.", + "fix": "Title detection needs improvement for preproof covers" + }, + { + "bug": "authors_not_classified", + "pages": [1], + "description": "Author list gets role=frontmatter_noise instead of authors.", + "fix": "Author detection needs improvement" + }, + { + "bug": "backmatter_heading_not_recognized", + "pages": [25, 35], + "description": "Conflict of Interest, Acknowledgments, Table and Figure Captions have role=section_heading/subsection_heading instead of backmatter_heading.", + "fix": "Backmatter headings need dedicated classification" + } ] } diff --git a/tests/fixtures/ocr_real_papers/regenerate_traces.py b/tests/fixtures/ocr_real_papers/regenerate_traces.py new file mode 100644 index 00000000..0c7f7c41 --- /dev/null +++ b/tests/fixtures/ocr_real_papers/regenerate_traces.py @@ -0,0 +1,130 @@ +"""Regenerate block_trace.csv by re-running build_structured_blocks from vault data.""" +from __future__ import annotations + +import csv +import json +import sys +from pathlib import Path + +sys.path.insert(0, r"D:\L\Med\Research\99_System\LiteraturePipeline\github-release\.worktrees\ocr-v2") + +from paperforge.worker.ocr_blocks import build_structured_blocks + +VAULT_OCR = Path(r"D:\L\OB\Literature-hub\System\PaperForge\ocr") +FIXTURE_ROOT = Path(r"D:\L\Med\Research\99_System\LiteraturePipeline\github-release\.worktrees\ocr-v2\tests\fixtures\ocr_real_papers") + +FIELD_NAMES = [ + "page", "block_id", "raw_label", "content_preview", "bbox", + "role", "role_confidence", "evidence", + "seed_role", "seed_confidence", + "zone", "style_family", "marker_type", + "render_default", "index_default", +] + + +def load_jsonl(path: Path) -> list[dict]: + return [json.loads(line) for line in path.read_text(encoding="utf-8").splitlines() if line.strip()] + + +def load_json(path: Path) -> dict: + return json.loads(path.read_text(encoding="utf-8")) + + +def regenerate(paper_key: str) -> None: + paper_dir = VAULT_OCR / paper_key + raw_jsonl = paper_dir / "canonical" / "blocks.raw.jsonl" + meta_json = paper_dir / "raw" / "source_metadata.json" + result_json = paper_dir / "json" / "result.json" + + if not raw_jsonl.exists(): + print(f"Skipping {paper_key}: no blocks.raw.jsonl at {raw_jsonl}") + return + + # Load raw blocks + raw_blocks = load_jsonl(raw_jsonl) + print(f"\n=== Regenerating {paper_key} ({len(raw_blocks)} raw blocks) ===") + + # Load source metadata + source_metadata = None + if meta_json.exists(): + try: + source_metadata = load_json(meta_json) + print(f" Source metadata loaded") + except Exception: + pass + + # Run build_structured_blocks + rows, doc_structure = build_structured_blocks(raw_blocks, source_metadata=source_metadata) + + if not rows: + print(f" ERROR: No rows returned") + return + + # Also load raw block text for fuller content_preview + raw_lookup: dict[tuple[int, int], str] = {} + try: + raw_data = load_json(result_json) + for payload in raw_data: + for res in payload.get("layoutParsingResults", []): + for block in res.get("prunedResult", {}).get("parsing_res_list", []): + page = block.get("page", 0) + bid = block.get("block_id", 0) + if page and bid: + raw_lookup[(page, bid)] = str(block.get("block_content", "")) + except Exception: + pass + + # Write block_trace.csv + out_path = FIXTURE_ROOT / paper_key / "block_trace.csv" + with open(out_path, "w", newline="", encoding="utf-8-sig") as f: + writer = csv.DictWriter(f, fieldnames=FIELD_NAMES) + writer.writeheader() + for row in rows: + page = row.get("page", 0) + bid = row.get("block_id", "") + bbox = row.get("bbox", [0, 0, 0, 0]) + + raw_content = raw_lookup.get((page, bid), row.get("text", "")) + content_preview = str(raw_content)[:100].replace("\n", " ") + + marker = row.get("marker_signature") or {} + writer.writerow({ + "page": page, + "block_id": bid, + "raw_label": row.get("raw_label", ""), + "content_preview": content_preview, + "bbox": str(bbox), + "role": row.get("role", ""), + "role_confidence": row.get("role_confidence", 0), + "evidence": "; ".join(row.get("evidence", [])[:3]) if row.get("evidence") else "", + "seed_role": row.get("seed_role", ""), + "seed_confidence": row.get("seed_confidence", 0), + "zone": row.get("zone", ""), + "style_family": row.get("style_family", ""), + "marker_type": marker.get("type", ""), + "render_default": row.get("render_default", False), + "index_default": row.get("index_default", False), + }) + + # Print role/zone distribution + role_counts: dict[str, int] = {} + zone_counts: dict[str, int] = {} + for row in rows: + role = row.get("role", "unassigned") + zone = row.get("zone", "") + role_counts[role] = role_counts.get(role, 0) + 1 + zone_counts[zone] = zone_counts.get(zone, 0) + 1 + + print(f" Wrote {len(rows)} rows to {out_path.name}") + print(f" Role distribution:") + for role, count in sorted(role_counts.items(), key=lambda x: -x[1]): + print(f" {role:40s} {count}") + print(f" Zone distribution:") + for zone, count in sorted(zone_counts.items(), key=lambda x: -x[1]): + print(f" {zone:40s} {count}") + + +if __name__ == "__main__": + for key in ["DWQQK2YB", "CAQNW9Q2"]: + regenerate(key) + print("\nDone. Run: python -m pytest tests/test_ocr_trace_vs_expectations.py -s") diff --git a/tests/test_ocr_trace_vs_expectations.py b/tests/test_ocr_trace_vs_expectations.py new file mode 100644 index 00000000..4862f5bb --- /dev/null +++ b/tests/test_ocr_trace_vs_expectations.py @@ -0,0 +1,346 @@ +#!/usr/bin/env python3 +""" +Compare real OCR pipeline output (block_trace.csv) against expectations.json. + +Usage: + python -m pytest tests/test_ocr_trace_vs_expectations.py -s --timeout=60 + python -m pytest tests/test_ocr_trace_vs_expectations.py -k "DWQQK2YB" -s + python -m pytest tests/test_ocr_trace_vs_expectations.py -k "CAQNW9Q2" -s + +Each test: + 1. Loads block_trace.csv from the real output directory + 2. Loads expectations.json from the test fixtures + 3. Runs all assertions defined in expectations.json + 4. Reports PASS/FAIL per assertion with details +""" + +from __future__ import annotations + +import csv +import json +import re +from pathlib import Path +from typing import Any + +import pytest + +# --------------------------------------------------------------------------- +# Config: map paper keys to fixture directory (contains block_trace.csv + expectations.json) +# --------------------------------------------------------------------------- + +FIXTURE_ROOT = Path(__file__).resolve().parent / "fixtures" / "ocr_real_papers" + +PAPER_CONFIGS = { + "DWQQK2YB": FIXTURE_ROOT / "DWQQK2YB", + "CAQNW9Q2": FIXTURE_ROOT / "CAQNW9Q2", +} + + +# --------------------------------------------------------------------------- +# Load block_trace.csv +# --------------------------------------------------------------------------- + + +def load_trace(trace_path: Path) -> list[dict]: + """Load block_trace.csv into a list of dicts.""" + with open(trace_path, "r", encoding="utf-8-sig") as f: + reader = csv.DictReader(f) + return list(reader) + + +def get_blocks_on_page(trace: list[dict], page: int) -> list[dict]: + """Get all blocks on a specific page.""" + return [b for b in trace if int(b.get("page", 0)) == page] + + +def find_block_matching(trace: list[dict], page: int, text_contains: str) -> dict | None: + """Find the first block on page whose content_preview contains the text.""" + blocks = get_blocks_on_page(trace, page) + for b in blocks: + cp = str(b.get("content_preview", "")).lower() + if text_contains.lower() in cp: + return b + return None + + +# --------------------------------------------------------------------------- +# Assertion runners +# --------------------------------------------------------------------------- + + +class AssertionResult: + def __init__(self, page: int, assertion_type: str, passed: bool, message: str, severity: str = "FAIL"): + self.page = page + self.assertion_type = assertion_type + self.passed = passed + self.message = message + self.severity = severity # FAIL or WARN + + def __repr__(self): + status = "PASS" if self.passed else self.severity + return f"[{status}] p{self.page} {self.assertion_type}: {self.message}" + + +def run_text_equals(trace: list[dict], page: int, text: str, expected_role: str | None = None, + expected_zone: str | None = None, confidence: str = "medium", + notes: str = "") -> AssertionResult: + """Assert a block with exact text match exists and has correct role/zone.""" + block = find_block_matching(trace, page, text) + if block is None: + return AssertionResult(page, "text_equals", False, + f"Block with text '{text[:50]}' NOT FOUND on page {page}", + severity="FAIL") + + errors = [] + if expected_role and block.get("role") != expected_role: + errors.append(f"role: '{block.get('role')}' != expected '{expected_role}'") + if expected_zone and block.get("zone") != expected_zone: + errors.append(f"zone: '{block.get('zone')}' != expected '{expected_zone}'") + + if errors: + return AssertionResult(page, "text_equals", False, + f"'{text[:40]}...' -> {'; '.join(errors)}", + severity="FAIL") + + return AssertionResult(page, "text_equals", True, + f"'{text[:50]}' role={block.get('role')} zone={block.get('zone')}") + + +def run_text_contains(trace: list[dict], page: int, text: str, expected_role: str | None = None, + expected_zone: str | None = None, must_not_role: str | None = None, + notes: str = "") -> AssertionResult: + """Assert a block containing text exists with correct role/zone.""" + block = find_block_matching(trace, page, text) + if block is None: + return AssertionResult(page, "text_contains", False, + f"Block containing '{text[:50]}' NOT FOUND on page {page}", + severity="FAIL") + + errors = [] + if expected_role and block.get("role") != expected_role: + errors.append(f"role: '{block.get('role')}' != expected '{expected_role}'") + if expected_zone and block.get("zone") != expected_zone: + errors.append(f"zone: '{block.get('zone')}' != expected '{expected_zone}'") + if must_not_role and block.get("role") == must_not_role: + errors.append(f"role must NOT be '{must_not_role}' but got '{block.get('role')}'") + + if errors: + return AssertionResult(page, "text_contains", False, + f"'{text[:40]}...' -> {'; '.join(errors)}", + severity="FAIL") + + return AssertionResult(page, "text_contains", True, + f"'{text[:50]}' role={block.get('role')} zone={block.get('zone')}") + + +def run_count_assertion(trace: list[dict], page: int, count_type: str, + min_count: int | None = None, max_count: int | None = None) -> AssertionResult: + """Assert count of a specific block type on a page.""" + blocks = get_blocks_on_page(trace, page) + + if count_type == "text": + count = sum(1 for b in blocks if b.get("raw_label") == "text") + elif count_type == "figure_title": + count = sum(1 for b in blocks if b.get("raw_label") in ("figure_title", "figure_caption") + or b.get("role") in ("figure_title", "figure_caption")) + elif count_type == "reference_content": + count = sum(1 for b in blocks if b.get("role") == "reference_item") + elif count_type == "image": + count = sum(1 for b in blocks if b.get("raw_label") == "image") + elif count_type == "chart": + # Charts often appear as image blocks with chart-like content + count = sum(1 for b in blocks if b.get("raw_label") == "image" + and any(kw in str(b.get("content_preview", "")).lower() + for kw in ["chart", "graph", "bar", "plot"])) + elif count_type == "all": + count = len(blocks) + else: + return AssertionResult(page, f"count_{count_type}", False, + f"Unknown count type: {count_type}", severity="FAIL") + + errors = [] + if min_count is not None and count < min_count: + errors.append(f"count {count} < min {min_count}") + if max_count is not None and count > max_count: + errors.append(f"count {count} > max {max_count}") + + if errors: + return AssertionResult(page, f"count_{count_type}", False, + f"{'; '.join(errors)} (actual={count})", + severity="WARN") + + return AssertionResult(page, f"count_{count_type}", True, + f"count={count} (range [{min_count or 0}, {max_count or 'inf'}])") + + +# --------------------------------------------------------------------------- +# Main test logic +# --------------------------------------------------------------------------- + + +def run_expectations_test(paper_key: str) -> list[AssertionResult]: + """Load trace + expectations, run all assertions, return results.""" + fixture_dir = PAPER_CONFIGS[paper_key] + trace_path = fixture_dir / "block_trace.csv" + expectations_path = fixture_dir / "expectations.json" + + assert trace_path.exists(), f"block_trace.csv not found: {trace_path}" + assert expectations_path.exists(), f"expectations.json not found: {expectations_path}" + + trace = load_trace(trace_path) + with open(expectations_path, "r", encoding="utf-8") as f: + expectations = json.load(f) + + results: list[AssertionResult] = [] + + # Run page-level assertions + pages_exp = expectations.get("pages", {}) + for page_str, page_exp in pages_exp.items(): + page_num = int(page_str) + for assertion in page_exp.get("assertions", []): + # text_equals assertion + if "text_equals" in assertion: + r = run_text_equals( + trace, page_num, assertion["text_equals"], + expected_role=assertion.get("expected_role"), + expected_zone=assertion.get("expected_zone"), + notes=assertion.get("notes", ""), + ) + results.append(r) + + # text_contains assertion + if "text_contains" in assertion: + r = run_text_contains( + trace, page_num, assertion["text_contains"], + expected_role=assertion.get("expected_role"), + expected_zone=assertion.get("expected_zone"), + must_not_role=assertion.get("must_not_role"), + notes=assertion.get("notes", ""), + ) + results.append(r) + + # count_text assertion + if "count_text" in assertion: + ct = assertion["count_text"] + r = run_count_assertion(trace, page_num, "text", + min_count=ct.get("min"), max_count=ct.get("max")) + results.append(r) + + # count_figure_title assertion + if "count_figure_title" in assertion: + ct = assertion["count_figure_title"] + r = run_count_assertion(trace, page_num, "figure_title", + min_count=ct.get("min"), max_count=ct.get("max")) + results.append(r) + + # count_reference_content assertion + if "count_reference_content" in assertion: + ct = assertion["count_reference_content"] + r = run_count_assertion(trace, page_num, "reference_content", + min_count=ct.get("min"), max_count=ct.get("max")) + results.append(r) + + # count_image assertion + if "count_image" in assertion: + ct = assertion["count_image"] + r = run_count_assertion(trace, page_num, "image", + min_count=ct.get("min"), max_count=ct.get("max")) + results.append(r) + + # count_chart assertion + if "count_chart" in assertion: + ct = assertion["count_chart"] + r = run_count_assertion(trace, page_num, "chart", + min_count=ct.get("min"), max_count=ct.get("max")) + results.append(r) + + # count_all assertion + if "count_all" in assertion: + ct = assertion["count_all"] + r = run_count_assertion(trace, page_num, "all", + min_count=ct.get("min"), max_count=ct.get("max")) + results.append(r) + + return results + + +def format_results_report(paper_key: str, results: list[AssertionResult]) -> str: + """Format results into a readable report.""" + lines = [] + lines.append("=" * 70) + lines.append(f"TRACE vs EXPECTATIONS REPORT: {paper_key}") + lines.append("=" * 70) + + passed = [r for r in results if r.passed] + failed = [r for r in results if not r.passed] + + lines.append(f"\nTotal: {len(results)} assertions | PASS: {len(passed)} | FAIL: {len(failed)}") + lines.append("") + + if failed: + lines.append("--- FAILURES ---") + by_page: dict[int, list[AssertionResult]] = {} + for r in failed: + by_page.setdefault(r.page, []).append(r) + for pg in sorted(by_page): + lines.append(f"\n Page {pg}:") + for r in by_page[pg]: + lines.append(f" [{r.severity}] {r.assertion_type}: {r.message}") + lines.append("") + + if passed: + lines.append(f"--- PASS ({len(passed)}) ---") + for r in passed[:20]: # Show first 20 passes + lines.append(f" [OK] p{r.page} {r.assertion_type}: {r.message}") + if len(passed) > 20: + lines.append(f" ... and {len(passed) - 20} more") + lines.append("") + + lines.append("=" * 70) + return "\n".join(lines) + + +# --------------------------------------------------------------------------- +# Pytest tests +# --------------------------------------------------------------------------- + + +@pytest.mark.parametrize("paper_key", ["DWQQK2YB", "CAQNW9Q2"]) +def test_trace_vs_expectations(paper_key: str) -> None: + """Compare real block_trace.csv against expectations.json for a paper.""" + results = run_expectations_test(paper_key) + report = format_results_report(paper_key, results) + + # Print report (will be captured by pytest -s) + print("\n" + report) + + # Fail test if any FAIL-level assertions failed + fails = [r for r in results if not r.passed and r.severity == "FAIL"] + if fails: + fail_summary = "\n".join(f" {r}" for r in fails) + pytest.fail(f"{len(fails)} FAIL-level assertions failed:\n{fail_summary}") + + +@pytest.mark.parametrize("paper_key", ["DWQQK2YB", "CAQNW9Q2"]) +def test_trace_role_distribution(paper_key: str) -> None: + """Print role distribution from real trace for manual review.""" + fixture_dir = PAPER_CONFIGS[paper_key] + trace_path = fixture_dir / "block_trace.csv" + if not trace_path.exists(): + pytest.skip(f"block_trace.csv not found: {trace_path}") + + trace = load_trace(trace_path) + from collections import Counter + role_counts = Counter(b.get("role", "?") for b in trace) + zone_counts = Counter(b.get("zone", "?") for b in trace) + + print(f"\n{paper_key} - Role Distribution:") + for role, count in role_counts.most_common(): + print(f" {role:35s} {count:4d}") + print(f"\n{paper_key} - Zone Distribution:") + for zone, count in zone_counts.most_common(): + print(f" {zone:35s} {count:4d}") + + +if __name__ == "__main__": + pytest.main([__file__, "-s", "-v", "--timeout=60"])