{"id":"W6977765125","doi":"10.6084/m9.figshare.c.7612879.v1","title":"External validation of AI-based scoring systems in the ICU: a systematic review and meta-analysis","year":2025,"lang":"en","type":"other","venue":"Figshare","topic":"COVID-19 Digital Contact Tracing","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Overfitting; Reliability (semiconductor); Receiver operating characteristic; External validity; Intensive care unit; Risk assessment; Intensive care","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0003697825,0.0002173101,0.001473666,0.0005072567,0.0000188947,0.0002972047,0.0008398579,0.00007587569,0.003477186],"category_scores_gemma":[0.00076914,0.0001412462,0.0004371396,0.001046738,0.000002672696,0.0001617502,0.0001405646,0.0001457499,0.00005164844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004262794,"about_ca_system_score_gemma":0.0001387401,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001855614,"about_ca_topic_score_gemma":0.00006289612,"domain_scores_codex":[0.9982634,0.000320937,0.0005165279,0.0003641813,0.000406872,0.0001281492],"domain_scores_gemma":[0.9978665,0.0005270974,0.0005395969,0.000954157,0.00008695645,0.00002573086],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"systematic_review","study_design_scores_codex":[3.205994e-7,0.00001938601,0.000004132709,0.9643663,0.007611341,0.00002635358,0.0000498111,0.0001918785,4.367546e-7,0.0003537245,0.02736465,0.00001169439],"study_design_scores_gemma":[0.0002545404,0.00004490296,0.00004567841,0.7620367,0.2175072,0.00001251233,0.00001626289,0.01068839,0.00005155548,0.00005085773,0.008471291,0.0008200703],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"other","genre_scores_codex":[3.896577e-7,0.7667847,0.04616169,0.001195501,0.0001240827,0.01118075,0.1216653,0.0004133972,0.05247424],"genre_scores_gemma":[0.2721079,0.0008054201,0.005464732,0.05107014,0.0005325695,0.03500087,0.07019661,0.001441437,0.5633803],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.7659793,"threshold_uncertainty_score":0.9974338,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09521250061295855,"score_gpt":0.3238378408534722,"score_spread":0.2286253402405136,"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."}}