{"id":"W4408039465","doi":"10.62617/mcb1326","title":"Weakly-supervised natural language processing with BERT-Clinical for automated lesion information extraction from free-text MRI reports in multiple sclerosis patients","year":2025,"lang":"en","type":"article","venue":"Molecular & cellular biomechanics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Information extraction; Computer science; Text messaging; Natural language processing; Artificial intelligence; Multiple sclerosis; Lesion; Medicine; Information retrieval; Pattern recognition (psychology); Pathology; World Wide Web","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":[0.003492386,0.001132417,0.0005665144,0.001138729,0.0002943773,0.0007483039,0.00102116,0.0009294748,0.001601179],"category_scores_gemma":[0.008533302,0.0003218824,0.001107931,0.0006265807,0.0003321505,0.0009496951,0.0007303481,0.00145879,0.0009466219],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009923044,"about_ca_system_score_gemma":0.001030836,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00551366,"about_ca_topic_score_gemma":0.007492333,"domain_scores_codex":[0.9985393,0.0008601664,0.0001008102,0.0002877681,0.0001268017,0.00008518741],"domain_scores_gemma":[0.9933878,0.005355409,0.0003673108,0.0003333425,0.0004618545,0.00009422954],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003815901,0.001473249,0.02978607,0.0009619527,0.000692526,0.001016685,0.001033989,0.2745493,0.01880847,0.002248629,0.01337656,0.6522367],"study_design_scores_gemma":[0.00005030077,0.0002647301,0.003745998,0.00003165512,0.00008361277,0.0001506342,0.00008909509,0.9890676,0.003494897,0.001794195,0.001202782,0.00002448655],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6195887,0.002903903,0.3535036,0.002044315,0.0002326533,0.0007816719,0.007789145,0.009913602,0.003242496],"genre_scores_gemma":[0.9161738,0.0002691544,0.07186443,0.0003045849,0.00009898775,0.0004109701,0.009214507,0.00009387654,0.001569714],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00551366,"threshold_uncertainty_score":0.01846969,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01565492868593291,"score_gpt":0.2736805255182077,"score_spread":0.2580255968322748,"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."}}