{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004980164,0.0002665758,0.0005301632,0.0003325058,0.0009560589,0.0005079216,0.001306697,0.0001664522,0.000004466741],"category_scores_gemma":[0.002363455,0.0002325709,0.00006472108,0.0004676133,0.00005145939,0.001008939,0.001944855,0.0006232553,8.24818e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008337206,"about_ca_system_score_gemma":0.001300732,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006069181,"about_ca_topic_score_gemma":0.0001454621,"domain_scores_codex":[0.9964887,0.0002083523,0.001354809,0.0004806936,0.0008165297,0.0006508689],"domain_scores_gemma":[0.9947429,0.003455049,0.0005377583,0.0007855813,0.0001804787,0.0002981965],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001705835,0.00001625131,0.0003148056,0.0005730761,0.00001788545,0.000002687632,0.00257177,0.38127,8.367301e-7,0.03363917,0.0000759167,0.5815006],"study_design_scores_gemma":[0.0006508537,0.0000714975,0.000008486743,0.001800493,0.00001221048,0.00002472972,0.0006388077,0.9547858,0.000001087783,0.04149415,0.0003142113,0.0001977128],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01225695,0.001615914,0.9845625,0.0001958931,0.0003174952,0.0006350942,0.00001230468,0.000150132,0.0002537348],"genre_scores_gemma":[0.1572318,0.00009581853,0.8409093,0.001609032,0.0000633671,0.00002000725,0.0000419647,0.00001569394,0.00001297712],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5813029,"threshold_uncertainty_score":0.9483965,"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."}}