{"id":"W4410940960","doi":"10.1101/2025.06.01.25328302","title":"NeuroDiscovery AI database: Comprehensive EHR dataset for Neurology","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centennial College","funders":"","keywords":"Database; Neurology; Computer science; Data science; Medicine; Psychiatry","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"],"consensus_categories":[],"category_scores_codex":[0.0004919412,0.0005073314,0.0006837411,0.0003467723,0.0002322746,0.0002558256,0.004260544,0.0003505695,0.00002120324],"category_scores_gemma":[0.0005993953,0.0005219775,0.0001784208,0.0003040037,0.0001009274,0.0002687207,0.007916182,0.002256382,0.0000491584],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003477155,"about_ca_system_score_gemma":0.0006359579,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004125869,"about_ca_topic_score_gemma":0.00007962994,"domain_scores_codex":[0.9956184,0.0007178605,0.0006246783,0.001963705,0.0004009216,0.0006744138],"domain_scores_gemma":[0.9938628,0.001193126,0.0003600125,0.004137373,0.0002598919,0.0001868079],"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":[0.0003590219,0.0004246545,0.0724972,0.01230801,0.0003132939,0.000822067,0.0007044729,0.04930917,0.0004587515,0.06118658,0.7783087,0.02330801],"study_design_scores_gemma":[0.000469338,0.000184189,0.0157133,0.0001529971,0.00003784076,0.00003400608,0.000002195143,0.2690592,0.00008198396,0.003366296,0.7103419,0.0005567162],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"commentary","genre_scores_codex":[0.06277659,0.001619054,0.7458901,0.1127816,0.02111697,0.004217789,0.05011959,0.001042154,0.0004361471],"genre_scores_gemma":[0.2687432,0.001613909,0.2106056,0.3663328,0.005550487,0.004031262,0.1397337,0.0004580621,0.002930922],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5352846,"threshold_uncertainty_score":0.9997232,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04689471311080777,"score_gpt":0.3590166925781474,"score_spread":0.3121219794673397,"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."}}