{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001971479,0.00108461,0.001079062,0.003831382,0.000717537,0.001995903,0.002528702,0.001594325,0.010609],"category_scores_gemma":[0.01142665,0.0004041666,0.00104135,0.006424139,0.0003700294,0.00163943,0.001950339,0.001457046,0.01623542],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001785724,"about_ca_system_score_gemma":0.003446679,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01687003,"about_ca_topic_score_gemma":0.01997565,"domain_scores_codex":[0.997582,0.000428901,0.0005998405,0.0006190586,0.0005487213,0.0002214879],"domain_scores_gemma":[0.9952698,0.001015332,0.000713105,0.001130764,0.001261988,0.0006089856],"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":[0.0004469121,0.0001357176,0.0136682,0.0009787912,0.000143876,0.0002635587,0.0001207168,0.001848157,0.00055528,0.002053669,0.9663126,0.01347262],"study_design_scores_gemma":[0.0009613852,0.0001971322,0.06084461,0.0007433189,0.0001712293,0.0009903432,0.0005171077,0.008946075,0.001775742,0.006582578,0.9180889,0.0001815922],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.002373363,0.0002302249,0.0006774499,0.0003626678,0.00004637241,0.0001289099,0.9939584,0.001034222,0.001188543],"genre_scores_gemma":[0.00347914,0.0001056256,0.001263075,0.0001447393,0.00002141861,0.00021374,0.9944974,0.00004946489,0.0002254246],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01687003,"threshold_uncertainty_score":0.03549063,"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."}}