{"id":"W4391426559","doi":"10.1016/j.euo.2024.01.004","title":"External Validation of a Digital Pathology-based Multimodal Artificial Intelligence Architecture in the NRG/RTOG 9902 Phase 3 Trial","year":2024,"lang":"en","type":"article","venue":"European Urology Oncology","topic":"AI in cancer detection","field":"Computer Science","cited_by":26,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University Health Centre","funders":"National Cancer Institute; Foundation Medicine; University of California, San Francisco; EMD Serono; Astellas Pharma; Endocyte; Advanced Accelerator Applications; NRG Oncology; Varian Medical Systems; Sanofi; Invitae; Bristol-Myers Squibb; AstraZeneca; Amgen; Pfizer","keywords":"Medicine; Hazard ratio; Confidence interval; Oncology; Internal medicine; Randomized controlled trial; Clinical trial; Proportional hazards model; Histopathology; Pathology","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":[0.001710904,0.0001926205,0.0002631909,0.0003708314,0.00007143274,0.00010594,0.00110685,0.0001452563,0.00003431692],"category_scores_gemma":[0.0002603969,0.0001511371,0.0001255285,0.0005592508,0.0003844156,0.0002123486,0.0001878969,0.0007100169,0.000160749],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008726622,"about_ca_system_score_gemma":0.0002329972,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005558568,"about_ca_topic_score_gemma":0.00001586559,"domain_scores_codex":[0.9964837,0.001716288,0.0006179988,0.0006177014,0.0002189287,0.0003453218],"domain_scores_gemma":[0.9982783,0.0009360134,0.0001802376,0.0005087254,0.00005004745,0.00004671172],"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":[0.006097445,0.000848818,0.00005869967,0.0000225521,0.00002313527,0.002039015,0.004144343,0.009046307,0.005580504,0.007315118,0.0001428111,0.9646813],"study_design_scores_gemma":[0.04531971,0.06485826,0.002098201,0.0001522684,0.00025807,0.004753251,0.0004438213,0.6262587,0.030424,0.1556024,0.06795437,0.001876978],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2912213,0.0001349012,0.7014527,0.003065994,0.001922055,0.0003864978,0.00001009226,0.0001270631,0.001679402],"genre_scores_gemma":[0.9941841,0.000005265137,0.004577745,0.0006142114,0.000542003,0.00003744038,0.000008463935,0.00001980151,0.00001097573],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9628043,"threshold_uncertainty_score":0.6163191,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04146474258499896,"score_gpt":0.3321157138798393,"score_spread":0.2906509712948404,"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."}}