{"id":"W4411910806","doi":"10.3171/2025.4.focus25225","title":"Synthetic neurosurgical data generation with generative adversarial networks and large language models:an investigation on fidelity, utility, and privacy","year":2025,"lang":"en","type":"article","venue":"Neurosurgical FOCUS","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Computer science; Fidelity; Synthetic data; Machine learning; Artificial intelligence; Generative adversarial network; Data mining; Deep learning","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":["metaepi_narrow","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.0007252782,0.0003109543,0.0003197151,0.0001463999,0.0003295477,0.0005469105,0.006462893,0.0002163637,0.000004747736],"category_scores_gemma":[0.002544289,0.0002543664,0.00002600782,0.0005116244,0.0003100773,0.001429755,0.03198002,0.0005661735,0.000001402716],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002772669,"about_ca_system_score_gemma":0.00007252578,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005401118,"about_ca_topic_score_gemma":0.00005205039,"domain_scores_codex":[0.9966334,0.000503591,0.0003762761,0.001640445,0.000416973,0.000429326],"domain_scores_gemma":[0.9927279,0.0003971575,0.0001124641,0.006519856,0.00006428811,0.0001783482],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001298042,0.001855152,0.01012689,0.0001840255,0.0002233203,0.001987154,0.001184149,0.01089358,0.006048623,0.4688954,0.0534672,0.4438365],"study_design_scores_gemma":[0.0009738987,0.0002467045,0.001091758,0.00003138407,0.00002103025,0.00005574495,0.00001076927,0.9686328,0.0006536662,0.02740894,0.0006385141,0.0002347706],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4776145,0.0003120056,0.5075487,0.01264093,0.0003774526,0.0004857089,0.00009733619,0.0005718294,0.0003515333],"genre_scores_gemma":[0.9798779,0.000138285,0.01889434,0.0008161744,0.000105819,0.00002203378,0.0001084652,0.00001715041,0.00001983819],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9577392,"threshold_uncertainty_score":0.9999909,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06266774018845506,"score_gpt":0.292911142205895,"score_spread":0.2302434020174399,"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."}}