{"id":"W7117120252","doi":"10.1016/j.patrec.2025.12.008","title":"Towards robust and reliable multi-modal 3D segmentation of multiple sclerosis lesions","year":2025,"lang":"en","type":"article","venue":"Pattern Recognition Letters","topic":"Multiple Sclerosis Research Studies","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Surgical Specialties (Canada)","funders":"Regione Lombardia","keywords":"Segmentation; Robustness (evolution); Pattern recognition (psychology); Reliability (semiconductor); Context (archaeology); Image segmentation; Deep learning; Merge (version control)","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":[],"consensus_categories":[],"category_scores_codex":[0.0002386342,0.0001768332,0.0003231076,0.0003195196,0.0001537673,0.00002797888,0.00006697902,0.00007248943,0.0001424201],"category_scores_gemma":[0.0003339064,0.0001652297,0.00008940554,0.0002653891,0.0001738947,0.0001397763,0.0001098092,0.0001999594,0.00002829109],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009826347,"about_ca_system_score_gemma":0.0000396469,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006484498,"about_ca_topic_score_gemma":0.00008464221,"domain_scores_codex":[0.9985959,0.0000750863,0.0003763102,0.0003546072,0.000312691,0.0002854068],"domain_scores_gemma":[0.9992166,0.0001713625,0.0001046731,0.0002009712,0.0002059359,0.00010047],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0001479875,0.0002893058,0.2067161,0.000590637,0.0002155676,0.000005319801,0.000412744,0.00005603986,0.3917928,3.432997e-7,0.003707371,0.3960658],"study_design_scores_gemma":[0.006504725,0.0001186497,0.8880448,0.001417866,0.0001745184,0.000004319591,0.0005900318,0.008551207,0.09415676,0.000005722865,0.0002059585,0.0002254438],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9498793,0.0001409513,0.04238742,0.006095413,0.0001675258,0.0008454797,0.0001419645,0.00007074996,0.0002711446],"genre_scores_gemma":[0.9771497,0.0008131912,0.01849476,0.003011999,0.00005846434,0.0001812982,0.0001853256,0.00002167139,0.00008358673],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6813287,"threshold_uncertainty_score":0.6737871,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.117354473663626,"score_gpt":0.3153338853316429,"score_spread":0.1979794116680168,"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."}}