{"id":"W3173393692","doi":"10.1136/bmjopen-2020-047709","title":"Developing a reporting guideline for artificial intelligence-centred diagnostic test accuracy studies: the STARD-AI protocol","year":2021,"lang":"en","type":"article","venue":"BMJ Open","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":312,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa; Ottawa Hospital","funders":"NIHR Oxford Biomedical Research Centre; National Institute for Health and Care Research; UK Research and Innovation; NIHR Imperial Biomedical Research Centre; Cancer Research UK; Nuclear Power Institute of China","keywords":"Checklist; Medicine; Protocol (science); Test (biology); Diagnostic accuracy; Delphi method; Medical education; Transparency (behavior); Guideline; Medical physics; Artificial intelligence; Alternative medicine; Psychology; Computer science; Pathology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.003030658,0.0001886931,0.000478256,0.00004743992,0.0005327691,0.0002226424,0.0002958691,0.00008871368,0.0001940256],"category_scores_gemma":[0.2700611,0.0001331635,0.0001144913,0.000481066,0.00008064825,0.0002098875,0.000319959,0.0001955371,0.00006001229],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002888615,"about_ca_system_score_gemma":0.004022944,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003544939,"about_ca_topic_score_gemma":0.000965331,"domain_scores_codex":[0.995412,0.0001084345,0.003254247,0.0004668145,0.0003238663,0.0004346574],"domain_scores_gemma":[0.9816985,0.01118104,0.00214987,0.001059598,0.003712068,0.0001989434],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0009265622,0.0007395372,0.01212299,0.001915311,0.0002515181,0.0002785172,0.01470847,0.0001063492,0.0006736938,0.007611942,0.1007921,0.859873],"study_design_scores_gemma":[0.0003073415,0.0009955156,0.0008927522,0.005330826,0.0002907062,0.0004180932,0.1591832,0.004768979,0.2117879,0.09359172,0.5215629,0.0008700832],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.00693328,0.0002265296,0.04006099,0.3181361,0.001529875,0.6321012,0.0000214759,0.0001197317,0.0008707679],"genre_scores_gemma":[0.07806556,0.00006382279,0.04858805,0.02192206,0.00338446,0.846348,0.0001669567,0.00008278184,0.0013783],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.8590029,"threshold_uncertainty_score":0.7360875,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7499802218592867,"score_gpt":0.646529031348602,"score_spread":0.1034511905106847,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}