{"id":"W4393131103","doi":"10.1038/s41598-024-57207-7","title":"An evaluation of the replicability of analyses using synthetic health data","year":2024,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"Children's Hospital of Eastern Ontario; University of Ottawa","funders":"Canadian Institutes of Health Research; Mitacs; Canada Research Chairs; CHEO Research Institute; Compute Canada; Natural Sciences and Engineering Research Council of Canada; Bill and Melinda Gates Foundation","keywords":"Replicate; Computer science; Synthetic data; Data mining; Confidence interval; Logistic regression; Metric (unit); Missing data; Statistics; Machine learning; Artificial intelligence; Mathematics","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1186848,0.001297202,0.001014936,0.001603043,0.0009031111,0.00220101,0.00204049,0.001613607,0.001432806],"category_scores_gemma":[0.3016285,0.0005924112,0.001768695,0.001312769,0.002709142,0.002180066,0.002323556,0.002036308,0.0002378731],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001483488,"about_ca_system_score_gemma":0.001956124,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002809854,"about_ca_topic_score_gemma":0.001506179,"domain_scores_codex":[0.9511919,0.03631569,0.002780662,0.00397144,0.005152284,0.0005880413],"domain_scores_gemma":[0.5192897,0.3931223,0.01377025,0.05427169,0.01842531,0.001120764],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00614763,0.000840136,0.06579357,0.001421489,0.002407378,0.001034957,0.001969338,0.770814,0.01371781,0.02882271,0.003115044,0.1039159],"study_design_scores_gemma":[0.0006077574,0.002577742,0.01322948,0.0004173193,0.0005901281,0.0007033307,0.0006774572,0.9127598,0.02467855,0.03875674,0.00484714,0.0001545854],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5353016,0.001991961,0.4508859,0.001580272,0.0004514645,0.002124086,0.001789334,0.001348679,0.004526667],"genre_scores_gemma":[0.8855225,0.0002604843,0.1112715,0.0002798837,0.0000558247,0.000796245,0.001153022,0.0001732716,0.0004872924],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8813152,"threshold_uncertainty_score":0.6276724,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3019057160476462,"score_gpt":0.4646173370680689,"score_spread":0.1627116210204227,"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."}}