{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.005625514,0.0006353491,0.001072478,0.0006989927,0.001754032,0.00301871,0.0006917909,0.0005754587,0.000005826805],"category_scores_gemma":[0.00682389,0.0006438163,0.000258527,0.0007257126,0.0005375499,0.003712478,0.0006156756,0.001347173,0.000004110816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003657968,"about_ca_system_score_gemma":0.0009822326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004610019,"about_ca_topic_score_gemma":0.00001438963,"domain_scores_codex":[0.9938393,0.0009216811,0.003075533,0.0007844474,0.0005078386,0.0008712467],"domain_scores_gemma":[0.9935136,0.003279724,0.001639977,0.0007708044,0.0005099689,0.0002859021],"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":[0.003209948,0.0001072413,0.01019504,0.009136722,0.0001679298,0.000004627831,0.01853956,0.00519512,0.00000582658,0.05192459,0.00005788946,0.9014555],"study_design_scores_gemma":[0.007871047,0.0006319665,0.002799184,0.004301617,0.0001163698,0.000039699,0.001801085,0.947439,0.00005180851,0.03401823,0.0003660552,0.0005640082],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4103427,0.0006497668,0.5847733,0.0005202387,0.001470117,0.001874424,0.0000241142,0.0001482519,0.0001970654],"genre_scores_gemma":[0.7839257,0.00006066736,0.2147476,0.0007257202,0.0003263085,0.0001332153,0.00004095947,0.0000263322,0.00001346461],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9422438,"threshold_uncertainty_score":0.9996013,"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."}}