{"id":"W4410448172","doi":"10.3233/shti250398","title":"How Useful Is Synthetic Data in Developing Predictive Models for Health?","year":2025,"lang":"en","type":"article","venue":"Studies in health technology and informatics","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Hyperparameter; Synthetic data; Computer science; Bivariate analysis; Univariate; Fidelity; Pairwise comparison; Artificial intelligence; Machine learning; Data mining; Generative model; Multivariate statistics; Generative grammar","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":[],"consensus_categories":[],"category_scores_codex":[0.01443578,0.0006273223,0.0005996504,0.0011117,0.000450883,0.002154811,0.001461029,0.001763993,0.001324798],"category_scores_gemma":[0.0930426,0.000398508,0.00064109,0.001188207,0.001976093,0.002760787,0.001065714,0.001671412,0.0003389402],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001073546,"about_ca_system_score_gemma":0.001021257,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002285368,"about_ca_topic_score_gemma":0.001870759,"domain_scores_codex":[0.9939793,0.004597424,0.0002306586,0.0004628924,0.0006002816,0.0001294212],"domain_scores_gemma":[0.9248896,0.06030015,0.003209608,0.008797882,0.002394358,0.0004083762],"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":[0.000340159,0.0002039617,0.03255435,0.000289137,0.000199982,0.0002177961,0.0003904907,0.8800406,0.001410478,0.04353473,0.003112889,0.0377055],"study_design_scores_gemma":[0.00004001606,0.0001701019,0.00330917,0.0001334149,0.00003159532,0.0002508184,0.0002644657,0.9202893,0.002198708,0.07027937,0.0029917,0.00004134087],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3020031,0.001477946,0.6784944,0.0068381,0.0003450848,0.0004261992,0.004246354,0.000615051,0.00555375],"genre_scores_gemma":[0.9415429,0.0004131488,0.05501602,0.00053369,0.00007983891,0.0001855636,0.001853058,0.00004880804,0.0003269313],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01443578,"threshold_uncertainty_score":0.07634461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1270629445431891,"score_gpt":0.3900612207030319,"score_spread":0.2629982761598428,"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."}}