{"id":"W4301370299","doi":"10.1136/bjo-2022-321842","title":"Artificial intelligence-enabled retinal vasculometry for prediction of circulatory mortality, myocardial infarction and stroke","year":2022,"lang":"en","type":"article","venue":"British Journal of Ophthalmology","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"Population Health Research Institute","funders":"Age UK; Medical Research Council; National Institute for Health and Care Research; British Heart Foundation; Cancer Research UK","keywords":"Medicine; Stroke (engine); Myocardial infarction; Internal medicine; Proportional hazards model; Cardiology; Cohort; Prospective cohort study; Framingham Risk Score; European Prospective Investigation into Cancer and Nutrition; Surgery; Disease","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":[0.001303054,0.00008442997,0.000466848,0.0003199625,0.0001539371,0.00001833506,0.00007987444,0.00006120164,0.000269398],"category_scores_gemma":[0.0003296517,0.0001002943,0.0003238095,0.0002362234,0.000141849,0.0000844314,0.00005242904,0.0003870922,2.798347e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006988845,"about_ca_system_score_gemma":0.0001177759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001000322,"about_ca_topic_score_gemma":3.608982e-7,"domain_scores_codex":[0.9984161,0.0001790374,0.000692478,0.0001665477,0.0003781576,0.00016763],"domain_scores_gemma":[0.9989775,0.00008799382,0.0004164389,0.00009772417,0.0003232036,0.00009718082],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001584023,0.000970331,0.8954384,0.00040118,0.002650015,0.004044289,0.0002819831,0.00133286,0.04461523,0.0002447575,0.0007610586,0.04767586],"study_design_scores_gemma":[0.000646422,0.002337532,0.734642,0.0001100833,0.001071702,0.2556113,0.0009974064,0.001169068,0.0008613094,0.002206343,0.0002385546,0.0001083079],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9963801,0.001106064,0.001302526,0.0001696261,0.000437769,0.0001367606,0.0001107939,0.000005664003,0.0003506935],"genre_scores_gemma":[0.9986963,0.0001162821,0.0006372952,0.00002481275,0.0003785104,0.000009492975,0.00002917654,0.00001314714,0.00009500695],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.251567,"threshold_uncertainty_score":0.4089884,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04476321523353274,"score_gpt":0.3146269922775672,"score_spread":0.2698637770440345,"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."}}