{"id":"W4416338326","doi":"10.18280/isi.300906","title":"A Hybrid Semantic–Rule-Based NLP Framework Integrating DFCI and MSKCC Approaches for Clinical Trial Matching Using UMLS and FAISS","year":2025,"lang":"","type":"article","venue":"Ingénierie des systèmes d information","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Unified Medical Language System; Matching (statistics); Clinical trial; MEDLINE","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01557416,0.001573442,0.002119982,0.01278655,0.001838762,0.008827361,0.003505564,0.003290939,0.00834503],"category_scores_gemma":[0.02674981,0.001287171,0.004866177,0.00536412,0.001655751,0.004501779,0.005384256,0.002338696,0.003529256],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003014409,"about_ca_system_score_gemma":0.009731825,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01614805,"about_ca_topic_score_gemma":0.02215557,"domain_scores_codex":[0.9907499,0.002909512,0.002272365,0.001384712,0.002351115,0.0003323795],"domain_scores_gemma":[0.9848043,0.01007973,0.0009002675,0.001798583,0.001975666,0.0004414718],"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":[0.0005025167,0.0005151706,0.00577176,0.003712735,0.00107015,0.002939204,0.002081065,0.1197052,0.01362445,0.3609632,0.02746628,0.4616482],"study_design_scores_gemma":[0.0001806031,0.0001273601,0.00103548,0.001126331,0.0006815474,0.001572892,0.0005246173,0.4527108,0.0146922,0.3237613,0.2033805,0.000206269],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001411407,0.0004037916,0.9827093,0.0008811296,0.00007242008,0.000587147,0.002911161,0.007948941,0.003074793],"genre_scores_gemma":[0.03122188,0.0004837137,0.9603605,0.0005375294,0.00005328414,0.0004557076,0.005160563,0.0005711054,0.001155732],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01614805,"threshold_uncertainty_score":0.08236498,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08890866593982516,"score_gpt":0.3646598457104329,"score_spread":0.2757511797706078,"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."}}