{"id":"W4413774835","doi":"10.1016/j.compbiomed.2025.110986","title":"Polyp segmentation in colonoscopy images using DeepLabV3++","year":2025,"lang":"en","type":"article","venue":"Computers in Biology and Medicine","topic":"Colorectal Cancer Screening and Detection","field":"Medicine","cited_by":10,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia, Okanagan Campus","funders":"","keywords":"Colonoscopy; Artificial intelligence; Computer science; Computer vision; Segmentation; Medicine; Internal medicine; Colorectal cancer","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":[0.0001888936,0.00007295643,0.0002187635,0.0003274182,0.00003158834,0.00000194237,0.00002590412,0.000084463,0.000007280913],"category_scores_gemma":[0.0000545691,0.00006048393,0.00001058369,0.0002931573,0.0001487447,0.00002201136,0.00002666321,0.0001403673,3.371086e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009142892,"about_ca_system_score_gemma":0.00003947736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005198882,"about_ca_topic_score_gemma":0.0001143422,"domain_scores_codex":[0.999455,0.00004853405,0.0001700509,0.0001792421,0.00002858643,0.0001186379],"domain_scores_gemma":[0.9997473,0.000108767,0.00002964311,0.00006630168,0.00001760501,0.00003037147],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.004490225,0.00006793062,0.856277,0.0001258695,0.00003996588,0.00004538184,0.0006486056,0.0000308418,0.04767413,0.0007931908,0.0006227189,0.08918416],"study_design_scores_gemma":[0.0120716,0.005570618,0.9465265,0.00191595,0.00009320474,0.0001382614,0.0007003461,0.01539962,0.01221755,0.004209432,0.0009466497,0.0002102487],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9707132,0.00233221,0.02368528,0.00165835,0.0006500424,0.0001913941,4.724295e-7,0.00002319911,0.0007458669],"genre_scores_gemma":[0.9932452,0.0001972176,0.005174465,0.001243277,0.00008043436,0.00000765478,0.00001047033,0.000002682284,0.00003860681],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.09024952,"threshold_uncertainty_score":0.2466463,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01925176810484388,"score_gpt":0.3726223900011238,"score_spread":0.35337062189628,"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."}}