{"id":"W3112106476","doi":"10.1038/s41598-020-79173-6","title":"Impact of data on generalization of AI for surgical intelligence applications","year":2020,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Surgical Simulation and Training","field":"Medicine","cited_by":89,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Workflow; Computer science; Context (archaeology); Artificial intelligence; Data science; Deep learning; Generalization","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":[0.0004145476,0.00004883206,0.0001532175,0.00005143442,0.0000389658,0.0000128691,0.00007141185,0.0000286644,0.0002104054],"category_scores_gemma":[0.0002189815,0.00003658577,0.00008188007,0.000418153,0.00008925075,0.00006021107,0.00003076807,0.00003901555,0.000002709568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001004131,"about_ca_system_score_gemma":0.0001208064,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005310766,"about_ca_topic_score_gemma":4.365953e-7,"domain_scores_codex":[0.9989655,0.00001054715,0.0003865107,0.0003210751,0.0002388516,0.00007750349],"domain_scores_gemma":[0.9989554,0.00007062228,0.000172592,0.0004905949,0.0002083607,0.0001024646],"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":[0.002093755,0.002673864,0.3462422,0.001055636,0.0005520596,0.0003088495,0.002447976,0.07709754,0.03359623,0.02561545,0.0229921,0.4853243],"study_design_scores_gemma":[0.003537129,0.001386698,0.02154334,0.0002525274,0.0002960165,0.0001903903,0.0001657501,0.395557,0.0705151,0.0100171,0.4960715,0.0004674659],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7987461,0.0001298251,0.1925956,0.001098871,0.0006266127,0.001829832,0.00009425297,0.00008362444,0.004795247],"genre_scores_gemma":[0.9984102,0.000001517985,0.0007599979,0.00003908808,0.00005716788,0.000008300444,0.0006232435,0.000005282043,0.00009521696],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4848569,"threshold_uncertainty_score":0.2303791,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1507180277890434,"score_gpt":0.4287067240858705,"score_spread":0.2779886962968271,"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."}}