{"id":"W3127841358","doi":"10.1038/s41598-021-95533-2","title":"A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography","year":2021,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Intracerebral and Subarachnoid Hemorrhage Research","field":"Medicine","cited_by":62,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; University of Toronto; St. Michael's Hospital","funders":"","keywords":"Generalizability theory; Computed tomography; Medical imaging; Neuroimaging; Head trauma; Tomography; Deep learning","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":[0.002725591,0.0001095445,0.000318022,0.0003265165,0.000144158,0.00004607096,0.00007221207,0.00006470932,0.00009385568],"category_scores_gemma":[0.0002448334,0.00007806531,0.000162001,0.001584342,0.0002339718,0.00007139756,0.00003407908,0.0003498984,0.00000146441],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005059882,"about_ca_system_score_gemma":0.0001866721,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004745446,"about_ca_topic_score_gemma":0.0009226669,"domain_scores_codex":[0.9975936,0.0003243802,0.0006947729,0.0004396868,0.0007279575,0.0002195823],"domain_scores_gemma":[0.9987315,0.00008960701,0.0002372746,0.0005567935,0.0003280655,0.0000567711],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002339182,0.000565541,0.05953258,0.0003259782,0.00001410838,0.001897943,0.0003853847,0.000121123,0.8581749,0.00007441657,0.00002181933,0.07865231],"study_design_scores_gemma":[0.001136622,0.0004729046,0.1086804,0.0001515086,0.00005689055,0.001750092,0.0002771249,0.1869886,0.6989395,0.001111705,0.0002541485,0.0001804678],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9960098,0.00008849191,0.001099875,0.0004257339,0.0006500725,0.0003892869,0.000002309911,0.00002165828,0.00131282],"genre_scores_gemma":[0.9960424,0.000009485382,0.003410253,0.0000241277,0.00005867902,0.00001230384,0.00007315219,0.00000768905,0.0003618878],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1868675,"threshold_uncertainty_score":0.3183411,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02138183746711629,"score_gpt":0.2942971536806372,"score_spread":0.2729153162135209,"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."}}