{"id":"W6977082170","doi":"10.6084/m9.figshare.25223503.v1","title":"Additional file 1 of Characterizing the limitations of using diagnosis codes in the context of machine learning for healthcare","year":2024,"lang":"en","type":"article","venue":"Figshare","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"SickKids Foundation","funders":"","keywords":"Context (archaeology); Health care; Key (lock); Training set","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003227346,0.0008585401,0.0009446254,0.003273423,0.0008559007,0.00226564,0.001758829,0.001802639,0.8914801],"category_scores_gemma":[0.121176,0.0005313162,0.001013771,0.003743519,0.0003395116,0.001969509,0.001251326,0.0009501544,0.1596257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001454082,"about_ca_system_score_gemma":0.002519551,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008864665,"about_ca_topic_score_gemma":0.01290397,"domain_scores_codex":[0.9982678,0.000471147,0.0003622614,0.0003916431,0.000353979,0.0001530788],"domain_scores_gemma":[0.8198831,0.1612031,0.004036158,0.004640061,0.009322416,0.0009151616],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003948809,0.0001455054,0.004478638,0.003585621,0.00007451126,0.0000727927,0.0001057139,0.0005667771,0.00006692338,0.001977568,0.9737797,0.0147513],"study_design_scores_gemma":[0.005655332,0.0005248258,0.06198664,0.01252526,0.0004798522,0.001489878,0.00150906,0.007353608,0.001416835,0.044719,0.8620623,0.0002774349],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"other","genre_scores_codex":[0.000368369,0.00003327823,0.0005332722,0.0001868739,0.00003399194,0.0001098211,0.9967339,0.0002190004,0.001781551],"genre_scores_gemma":[0.02508361,0.0002893255,0.01206744,0.001710307,0.0002270832,0.002960142,0.9421224,0.00128286,0.01425682],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.8914801,"threshold_uncertainty_score":0.1547904,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.470253144846657,"score_gpt":0.4764973894160241,"score_spread":0.006244244569367063,"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."}}