{"id":"W4307644840","doi":"10.1016/j.xops.2022.100235","title":"Use of Machine Learning to Assess Cataract Surgery Skill Level With Tool Detection","year":2022,"lang":"en","type":"article","venue":"Ophthalmology Science","topic":"Intraocular Surgery and Lenses","field":"Medicine","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University; Kingston Health Sciences Centre; Queen's University","funders":"","keywords":"Receiver operating characteristic; Cataract surgery; Medicine; Artificial intelligence; Confidence interval; Support vector machine; Machine learning; Medical physics; Computer science; Surgery; Internal medicine","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":[],"consensus_categories":[],"category_scores_codex":[0.001287261,0.00009905528,0.0002576481,0.0003572602,0.0004323215,0.00001477311,0.0001156091,0.00002943888,0.0003867019],"category_scores_gemma":[0.0009661665,0.00008303154,0.00005502857,0.001016825,0.0002962557,0.0002196598,0.0001055103,0.0003084153,0.000009739964],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007696515,"about_ca_system_score_gemma":0.0003559874,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004007442,"about_ca_topic_score_gemma":0.000004333332,"domain_scores_codex":[0.9986245,0.000123144,0.0001938242,0.0003294769,0.0004271067,0.0003019586],"domain_scores_gemma":[0.9989761,0.0004298961,0.0001068724,0.0002507133,0.0001471214,0.00008928031],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0008644479,0.0002774791,0.7739519,0.00003598766,0.00003469805,0.002333396,0.0005226102,0.00291571,0.2123411,0.00008403138,0.00005060841,0.006588108],"study_design_scores_gemma":[0.0001828059,0.001322665,0.8646798,0.00003429307,0.00003733743,0.02116358,0.0002524863,0.001036724,0.1086415,0.000011005,0.002410861,0.0002269973],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9986837,0.00002111005,0.0003646953,0.0001843141,0.000296894,0.0001484876,0.00001050611,0.00002045318,0.0002698555],"genre_scores_gemma":[0.9976944,0.000001478611,0.001035399,0.0001897319,0.00002123755,0.00002391841,0.000009647581,0.000009970977,0.001014184],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1036996,"threshold_uncertainty_score":0.4234115,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1159635794677132,"score_gpt":0.3157044724094968,"score_spread":0.1997408929417836,"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."}}