{"id":"W6902342541","doi":"10.6084/m9.figshare.29079400","title":"Additional file 1 of Investigating use of diagnostic codes for post-COVID- 19 condition in Ontario health administrative data","year":2025,"lang":"en","type":"article","venue":"Open MIND","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; University of Toronto; Institute for Clinical Evaluative Sciences; Sunnybrook Health Science Centre","funders":"","keywords":"Diagnosis code; Data collection; Confidentiality; Health data; MEDLINE; Public health; Identification (biology)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00106835,0.00007558474,0.0002954381,0.0001101285,0.0003114185,0.00001071801,0.0002694734,0.0001054943,0.3734274],"category_scores_gemma":[0.04695514,0.00007184923,0.00001656688,0.0001463823,0.00008987203,0.0003760455,0.0001794478,0.0003202897,0.00003949198],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003400288,"about_ca_system_score_gemma":0.01703108,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.03395594,"about_ca_topic_score_gemma":0.3871566,"domain_scores_codex":[0.9981868,0.0002816666,0.0009844184,0.0001821521,0.0001486745,0.0002163484],"domain_scores_gemma":[0.967203,0.03142394,0.0007353572,0.0002678849,0.0001864825,0.00018336],"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":[0.00006754498,0.0000531588,0.002363915,0.0006134873,0.000009194985,3.235512e-7,0.004006876,0.000007865384,0.000001852747,0.0003392737,0.9884683,0.004068174],"study_design_scores_gemma":[0.0008499267,0.0002345238,0.03477948,0.004453211,0.000008362578,3.508144e-7,0.003449197,0.0009426492,0.00001028766,0.0005243306,0.9546789,0.00006878866],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.01873738,0.000009846816,0.0004620089,0.004645954,0.0001306608,0.002946258,0.9673727,0.000002770885,0.005692461],"genre_scores_gemma":[0.03379559,0.000007009805,0.07708686,0.009399039,0.00004958057,0.001550602,0.8719889,0.000009266147,0.006113151],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.3733879,"threshold_uncertainty_score":0.9885414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6186593803172743,"score_gpt":0.5554088797517469,"score_spread":0.06325050056552739,"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."}}