{"id":"W4415902225","doi":"10.1016/j.mlwa.2025.100789","title":"Beyond single-run metrics with CP-fuse: A rigorous multi-cohort evaluation of clinico-pathological fusion for improved survival prediction in TCGA","year":2025,"lang":"en","type":"article","venue":"Machine Learning with Applications","topic":"AI in cancer detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke; McGill University; Jewish General Hospital","funders":"Canada Research Chairs","keywords":"Benchmark (surveying); Fusion; Survival analysis; Calibration; Pattern recognition (psychology); Feature (linguistics)","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002214977,0.0001901053,0.0002888258,0.0004511037,0.0002578138,0.00006287893,0.0004029878,0.0001421299,0.000004179943],"category_scores_gemma":[0.0003429455,0.0001565093,0.00005187952,0.002147703,0.00008607568,0.0002175927,0.0001151661,0.0004067377,0.000001820641],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003279648,"about_ca_system_score_gemma":0.0002201907,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001787911,"about_ca_topic_score_gemma":0.0005435771,"domain_scores_codex":[0.9978832,0.000258141,0.0004813747,0.00068955,0.0004445581,0.000243174],"domain_scores_gemma":[0.9980026,0.0004130289,0.000376942,0.0005618305,0.0005953477,0.00005021502],"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":[0.0004243917,0.00122141,0.4431386,0.000117768,0.0001066149,8.495002e-7,0.0004690502,0.07435539,0.009728977,0.005052253,0.00003368118,0.465351],"study_design_scores_gemma":[0.002890867,0.0008271174,0.1332001,0.00004274325,0.000131183,0.000005578946,0.00004675837,0.8595055,0.001087562,0.00101996,0.001064934,0.0001777058],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06725395,0.000145445,0.9287612,0.0003593193,0.0001209908,0.002401785,0.00001679486,0.0002127373,0.0007277829],"genre_scores_gemma":[0.8490344,0.00001643918,0.1484609,0.0000464943,0.00004226899,0.002147333,0.00007235676,0.00001804654,0.0001617881],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7851501,"threshold_uncertainty_score":0.6382263,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0315292332960606,"score_gpt":0.3145519740706642,"score_spread":0.2830227407746035,"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."}}