{"id":"W4389061858","doi":"10.1200/cci.23.00116","title":"Evaluating the Utility and Privacy of Synthetic Breast Cancer Clinical Trial Data Sets","year":2023,"lang":"en","type":"article","venue":"JCO Clinical Cancer Informatics","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University; McMaster University; Alberta Health Services; Ottawa Hospital; University of Ottawa; Agricultural Research Institute of Ontario","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Canadian Institutes of Health Research; Government of Ontario","keywords":"Computer science; Synthetic data; Data sharing; Generative model; Clinical trial; Data mining; Data type; Breast cancer; Machine learning; Artificial intelligence; Generative grammar; Medicine; Cancer; Alternative medicine","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0687916,0.0006173328,0.000622706,0.001135311,0.0004821316,0.001922812,0.00109751,0.001261138,0.001229944],"category_scores_gemma":[0.2094042,0.0003902183,0.001286245,0.0009763039,0.002246449,0.001571762,0.001949433,0.001731451,0.0001733072],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001975861,"about_ca_system_score_gemma":0.002106645,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001667585,"about_ca_topic_score_gemma":0.001148684,"domain_scores_codex":[0.9629936,0.03059376,0.001345915,0.002007494,0.002779542,0.0002796435],"domain_scores_gemma":[0.6316254,0.3300721,0.01028119,0.02237156,0.004824391,0.000825399],"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":[0.002702201,0.0003823797,0.05820055,0.0007666301,0.0009568011,0.000205421,0.0007858776,0.8031057,0.002930797,0.03209936,0.001926327,0.09593804],"study_design_scores_gemma":[0.0003500769,0.001197142,0.01169654,0.0002550672,0.0002355201,0.0003255878,0.0002843487,0.9163133,0.008606788,0.05752743,0.003132851,0.00007539125],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7254201,0.00131308,0.2628597,0.002995767,0.0001207105,0.0008850665,0.002602492,0.0003964613,0.003406667],"genre_scores_gemma":[0.9476436,0.0001655507,0.04987315,0.0003081791,0.0000252698,0.0003163943,0.001318161,0.00003094931,0.0003186759],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0687916,"threshold_uncertainty_score":0.3638091,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5048688417325166,"score_gpt":0.5517641226211488,"score_spread":0.04689528088863226,"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."}}