{"id":"W4412443577","doi":"10.1016/j.jval.2025.04.023","title":"P16 The Impact of Hallucinations in Synthetic Health Data on Prognostic Machine Learning Models","year":2025,"lang":"en","type":"article","venue":"Value in Health","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Children's Hospital of Eastern Ontario","funders":"","keywords":"Computer science; Machine learning; Artificial intelligence; Psychology","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":[0.005442759,0.0001974909,0.0004634987,0.0006342779,0.0002550861,0.00005174386,0.001916963,0.000066818,0.000005157145],"category_scores_gemma":[0.001722561,0.0001457934,0.00005832549,0.001731442,0.000050444,0.000260245,0.0005521217,0.001156236,0.000005460096],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006906422,"about_ca_system_score_gemma":0.002699137,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.1240399,"about_ca_topic_score_gemma":0.001609809,"domain_scores_codex":[0.9947597,0.00254661,0.0009886541,0.000673368,0.0003994396,0.0006322004],"domain_scores_gemma":[0.9957088,0.001993851,0.0004317927,0.001672613,0.00006266247,0.0001302787],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001058624,0.0001737957,0.05288385,0.0003602992,0.00000708361,0.00000278325,0.001990119,0.8500856,4.477225e-7,0.06276101,0.0001069806,0.03161745],"study_design_scores_gemma":[0.0002783704,0.0004370925,0.07718691,0.0009854552,9.255083e-7,0.00000445955,0.00004156718,0.9130757,5.902328e-7,0.007810256,0.00009585529,0.00008282877],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2977275,0.022648,0.332924,0.3320919,0.00137345,0.008235759,0.000215617,0.0007451244,0.004038688],"genre_scores_gemma":[0.9944383,0.0003610914,0.004330129,0.0007192949,0.00001704414,0.00003430478,0.00002741109,0.00001534995,0.00005707189],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6967108,"threshold_uncertainty_score":0.8817933,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1110057940299913,"score_gpt":0.3915406407849467,"score_spread":0.2805348467549553,"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."}}