{"id":"W4416786152","doi":"10.1186/s12911-025-03266-3","title":"Augmenting small tabular health data for training prognostic ensemble machine learning models using generative models","year":2025,"lang":"en","type":"article","venue":"BMC Medical Informatics and Decision Making","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; Ottawa Hospital; Children's Hospital of Eastern Ontario; Hospital for Sick Children; University of Ottawa","funders":"","keywords":"Categorical variable; Gradient boosting; Generative model; Decision tree; Pooling; Predictive modelling; Proxy (statistics); Boosting (machine learning); Ensemble learning; Sample (material)","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":[0.004828326,0.00101821,0.001281928,0.000807047,0.0003206344,0.001069403,0.00153919,0.001252362,0.001661611],"category_scores_gemma":[0.01102199,0.0006103852,0.001906028,0.0006563056,0.0009084287,0.001246039,0.001072957,0.002217524,0.0005212647],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007588142,"about_ca_system_score_gemma":0.0007704712,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003580374,"about_ca_topic_score_gemma":0.004022144,"domain_scores_codex":[0.9989622,0.0005365268,0.00005959826,0.0002475182,0.000119014,0.00007531822],"domain_scores_gemma":[0.9925668,0.005543034,0.000451755,0.0008042504,0.000518309,0.0001158446],"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":[0.0002550806,0.0001174747,0.01017564,0.0001067897,0.0001349311,0.0001002764,0.0001147068,0.9337508,0.001780248,0.002578543,0.001339241,0.04954623],"study_design_scores_gemma":[0.000009786213,0.00004624507,0.0007904058,0.00002085936,0.00002515534,0.00002553104,0.00001061338,0.9954143,0.0006630992,0.002663638,0.0003182207,0.00001211159],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1711036,0.0013529,0.821917,0.0009362219,0.0001393229,0.0002088681,0.00133087,0.001586065,0.001425175],"genre_scores_gemma":[0.8864869,0.0004407068,0.1090481,0.0004251004,0.00009857366,0.0003484161,0.00204755,0.00008374243,0.001020825],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004828326,"threshold_uncertainty_score":0.02553487,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2505484922964605,"score_gpt":0.4045887836879865,"score_spread":0.1540402913915259,"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."}}