{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001716772,0.0006183962,0.0008813076,0.003938551,0.001818475,0.001276177,0.001326477,0.0008266455,0.6567027],"category_scores_gemma":[0.04080036,0.0005785029,0.0008607112,0.007814321,0.0003950948,0.001002637,0.001103601,0.0006224345,0.04043576],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00713515,"about_ca_system_score_gemma":0.01551825,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.6034744,"about_ca_topic_score_gemma":0.7221126,"domain_scores_codex":[0.998136,0.0002525764,0.000355774,0.0003408245,0.0005497224,0.0003650141],"domain_scores_gemma":[0.9617784,0.01937536,0.004033872,0.002040208,0.01118884,0.001583351],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.0001070605,0.00005172962,0.009479932,0.0009977272,0.00003269168,0.00003080404,0.0001238961,0.0001410088,0.00004234194,0.0004602655,0.9846,0.003932516],"study_design_scores_gemma":[0.002114756,0.0001660452,0.2661043,0.003745391,0.0002027046,0.0003286786,0.001521094,0.001206718,0.0003940564,0.002933245,0.7211532,0.0001298408],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.0003037199,0.000008981331,0.00006187315,0.00005452019,0.000009888824,0.00006484226,0.9982324,0.00003229869,0.001231516],"genre_scores_gemma":[0.01585791,0.0001233476,0.001558858,0.0003586193,0.00007922195,0.001447227,0.9685599,0.0001823896,0.01183262],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6567027,"threshold_uncertainty_score":0.7977216,"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."}}