{"id":"W4392411623","doi":"10.2196/53241","title":"Creating High-Quality Synthetic Health Data: Framework for Model Development and Validation","year":2024,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Tensor decomposition and applications","field":"Mathematics","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Children's Hospital of Eastern Ontario; University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Quality (philosophy); Computer science; Model validation; Data quality; Data science; Data mining; Engineering; Operations management","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":true,"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":[0.002973024,0.00009907831,0.0001770992,0.0001537395,0.0005929141,0.0002301775,0.000220907,0.00006364262,0.00001930839],"category_scores_gemma":[0.0003062036,0.0000842848,0.00002361624,0.0002663672,0.00007350097,0.0003590593,0.0002069402,0.0003216073,0.0000399295],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001644032,"about_ca_system_score_gemma":0.0002161885,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008605577,"about_ca_topic_score_gemma":0.000004048583,"domain_scores_codex":[0.9983782,0.0001751784,0.0004167843,0.0003018064,0.0004149985,0.0003130422],"domain_scores_gemma":[0.9969603,0.002279507,0.00007074042,0.0004119293,0.0001669198,0.0001105836],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00001030495,0.00009470518,0.000003058919,0.0008991965,0.00002518434,1.64559e-7,0.01034918,0.000009588987,0.00007831346,0.9632716,0.004773046,0.0204857],"study_design_scores_gemma":[0.0001794598,0.00005760915,0.0001961648,0.0005145635,0.000005345998,0.000003558989,0.001741515,0.07783832,0.001998853,0.9115877,0.005724833,0.0001521403],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1468775,0.0001181433,0.8463427,0.004157036,0.00003780115,0.001261268,0.0003066142,0.0001681604,0.0007307685],"genre_scores_gemma":[0.733707,0.00002355497,0.2650491,0.00004547576,0.0000453216,0.0006500864,0.0001985887,0.0000205998,0.0002603616],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5868295,"threshold_uncertainty_score":0.4560275,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3724077758246818,"score_gpt":0.5654031685451946,"score_spread":0.1929953927205128,"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."}}