{"id":"W2946638670","doi":"10.1016/j.compmedimag.2019.04.006","title":"Model-free prostate cancer segmentation from dynamic contrast-enhanced MRI with recurrent convolutional networks: A feasibility study","year":2019,"lang":"en","type":"article","venue":"Computerized Medical Imaging and Graphics","topic":"Prostate Cancer Diagnosis and Treatment","field":"Medicine","cited_by":19,"is_retracted":false,"has_abstract":false,"ca_institutions":"Capital District Health Authority; Health Research Foundation; Nova Scotia Health Authority; Izaak Walton Killam Health Centre; Dalhousie University","funders":"Research Nova Scotia","keywords":"Computer science; Segmentation; Dynamic contrast; Magnetic resonance imaging; Convolutional neural network; Artificial intelligence; Dynamic contrast-enhanced MRI; Prostate cancer; Image segmentation; Deep learning; Contrast (vision); Pattern recognition (psychology); Computer vision; Cancer; Medicine; Radiology","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":[0.0002442825,0.0002586149,0.0004930365,0.00006884394,0.00008521687,0.00005145918,0.000110621,0.00005873611,0.00005012832],"category_scores_gemma":[0.0000214834,0.0001886003,0.0000662326,0.0001824321,0.0002181501,0.0001136086,0.000105193,0.0003230641,0.000001479377],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001576522,"about_ca_system_score_gemma":0.0002869296,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001388226,"about_ca_topic_score_gemma":0.0001940415,"domain_scores_codex":[0.9980291,0.00008323863,0.0003446636,0.000629332,0.000594197,0.0003194898],"domain_scores_gemma":[0.9988349,0.0001306418,0.0001286996,0.0003657108,0.0001794969,0.0003605247],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003477599,0.002844079,0.919387,0.0001816376,0.0008451497,0.0001274112,0.001874465,0.001209555,0.0005995827,0.0001132524,0.000681014,0.06865919],"study_design_scores_gemma":[0.02097644,0.0007421909,0.2012619,0.0008002662,0.0002816273,0.00001363476,0.0002065131,0.7748373,0.00003532511,0.0005529673,0.00005460271,0.0002372481],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9326831,0.003602494,0.05810334,0.003017725,0.0004134247,0.001956399,0.0001179891,0.0000931602,0.00001237867],"genre_scores_gemma":[0.9932163,0.002895235,0.002100813,0.001137941,0.00007766893,0.0002717997,0.0002589923,0.00002359684,0.00001759532],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7736277,"threshold_uncertainty_score":0.7690896,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01317546391848705,"score_gpt":0.2949795890372595,"score_spread":0.2818041251187725,"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."}}