{"id":"W4382654846","doi":"10.1200/cci.23.00071","title":"Status of Synthetic Data Generation for Structured Health Data","year":2023,"lang":"en","type":"editorial","venue":"JCO Clinical Cancer Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Agricultural Research Institute of Ontario; University of Ottawa","funders":"","keywords":"Synthetic data; Computer science; Data science; Artificial intelligence","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":["metaresearch","metaepi_narrow","open_science"],"consensus_categories":[],"category_scores_codex":[0.007378988,0.0003630963,0.001195847,0.0001629381,0.0002234849,0.000249208,0.00822803,0.0008450773,0.00001490093],"category_scores_gemma":[0.03494279,0.000327122,0.0001176715,0.000426865,0.0001142723,0.001071879,0.004465295,0.001984373,0.00001779398],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002496226,"about_ca_system_score_gemma":0.008281456,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001559142,"about_ca_topic_score_gemma":0.002246945,"domain_scores_codex":[0.9923443,0.0004538401,0.003926758,0.0008884051,0.001523957,0.0008627403],"domain_scores_gemma":[0.9778293,0.008967632,0.003431761,0.008550151,0.0007961139,0.0004250041],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00001582014,0.00001625184,0.00007486318,0.002079939,0.00005112851,2.945443e-7,0.000396218,0.0004403602,3.357678e-8,0.0001634271,0.867434,0.1293277],"study_design_scores_gemma":[0.0003369028,0.000175833,0.0000339493,0.0001790002,0.00002607222,1.61708e-7,0.00001914547,0.4542974,2.418516e-7,0.0001110918,0.5446529,0.0001672416],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"editorial","genre_gemma":"editorial","genre_scores_codex":[0.00001000593,0.0006114541,0.1339763,0.001181777,0.8412341,0.0009632103,0.02178161,0.0002101632,0.00003130106],"genre_scores_gemma":[0.00009036306,0.009458304,0.1996854,0.0007920974,0.7433817,0.00009737623,0.04621749,0.0001007605,0.0001765215],"genre_candidate":"editorial","genre_consensus":"editorial","teacher_disagreement_score":0.4538571,"threshold_uncertainty_score":0.9999181,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3509988297603718,"score_gpt":0.5450554057956647,"score_spread":0.1940565760352929,"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."}}