{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.006127438,0.0007460542,0.0005737965,0.0005110713,0.00037428,0.001533037,0.0005665535,0.001161253,0.006816901],"category_scores_gemma":[0.0632759,0.0003343946,0.0006483183,0.0004819629,0.0008022231,0.001386939,0.0009509809,0.001541108,0.0008717335],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006443245,"about_ca_system_score_gemma":0.0009374865,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008658704,"about_ca_topic_score_gemma":0.004564803,"domain_scores_codex":[0.9989063,0.0006092262,0.00007584484,0.0001513467,0.0001887232,0.00006855953],"domain_scores_gemma":[0.9511098,0.04347425,0.0008865853,0.001410826,0.00265533,0.0004630906],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002483416,0.0002118735,0.03614942,0.0004508062,0.000337655,0.0005673443,0.0001916683,0.8113224,0.002345205,0.005623323,0.009944908,0.1303719],"study_design_scores_gemma":[0.00004080113,0.0001522464,0.003180356,0.00005807501,0.00003343042,0.0001076923,0.0000454763,0.9895894,0.001784426,0.004471331,0.0005205869,0.00001627799],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7692886,0.002642332,0.1993357,0.01114684,0.001677244,0.0001496235,0.003663489,0.001449757,0.01064647],"genre_scores_gemma":[0.9890618,0.0002390091,0.008002526,0.000217755,0.0001128514,0.00001640419,0.0009516607,0.00006737099,0.001330513],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9938726,"threshold_uncertainty_score":0.03240532,"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."}}