{"id":"W4407633965","doi":"10.1007/s12672-025-01973-x","title":"Benchmarking histopathology foundation models for ovarian cancer bevacizumab treatment response prediction from whole slide images","year":2025,"lang":"en","type":"article","venue":"Discover Oncology","topic":"AI in cancer detection","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia Hospital; Vancouver General Hospital; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; BC Cancer Foundation; Canadian Institutes of Health Research; Canada's Michael Smith Genome Sciences Centre","keywords":"Benchmarking; Bevacizumab; Histopathology; Ovarian cancer; Foundation (evidence); Medicine; Medical physics; Computer science; Oncology; Cancer; Internal medicine; Pathology; Chemotherapy; Geography; Business","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":[0.0003635049,0.0001908958,0.0002945104,0.0002253759,0.0002701868,0.00009352175,0.0003493928,0.0001939251,0.00002482623],"category_scores_gemma":[0.00005143861,0.0001885481,0.0001045067,0.000258428,0.00007760793,0.0008253589,0.0001350596,0.000108955,0.000008894911],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003769775,"about_ca_system_score_gemma":0.001023458,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001368906,"about_ca_topic_score_gemma":0.0009080013,"domain_scores_codex":[0.9982111,0.0003118103,0.0003443553,0.000700579,0.0001244031,0.000307779],"domain_scores_gemma":[0.9986459,0.0005065534,0.000187618,0.0005082868,0.000105992,0.00004566618],"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":[0.003498937,0.0003525877,0.001576037,0.00002557327,0.000169049,0.00002872323,0.003483707,0.02553069,0.04264166,0.01495936,0.00584173,0.9018919],"study_design_scores_gemma":[0.006476574,0.002975592,0.04068736,0.00008861387,0.0002481944,0.00002082402,0.0001714979,0.5224315,0.01530765,0.09778252,0.3133299,0.000479775],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05161754,0.0003116686,0.9359947,0.005913467,0.004197939,0.0006008302,0.0001544603,0.0001467081,0.001062698],"genre_scores_gemma":[0.9559775,0.0001462368,0.03859552,0.0008435988,0.0004666884,0.00163211,0.0001271031,0.00002099926,0.002190232],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.90436,"threshold_uncertainty_score":0.9857831,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02478473114011203,"score_gpt":0.323553310106249,"score_spread":0.298768578966137,"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."}}