{"id":"W3036832166","doi":"10.1038/s41598-020-66849-2","title":"Histologic tissue components provide major cues for machine learning-based prostate cancer detection and grading on prostatectomy specimens","year":2020,"lang":"en","type":"article","venue":"Scientific Reports","topic":"AI in cancer detection","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"Lawson Health Research Institute; Western University","funders":"Canadian Institutes of Health Research; Prostate Cancer Canada; Government of Canada","keywords":"Grading (engineering); Prostatectomy; Prostate cancer; Artificial intelligence; Computer science; Transfer of learning; Cancer detection; Deep learning; Digital pathology; Pattern recognition (psychology); Prostate; Machine learning; Cancer; Medicine","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":[0.0006916989,0.00022665,0.0002409133,0.0001893862,0.0007325157,0.0004909806,0.000219852,0.00006326196,0.00001181934],"category_scores_gemma":[0.000184252,0.0002057038,0.00005900743,0.0006156038,0.0001579447,0.0003905131,0.0001098258,0.0002267157,0.000008045828],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002661314,"about_ca_system_score_gemma":0.0001238833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001203275,"about_ca_topic_score_gemma":0.0001396777,"domain_scores_codex":[0.9972789,0.00008216939,0.0004319807,0.001338209,0.0004750151,0.0003937735],"domain_scores_gemma":[0.9986658,0.00006567274,0.0004805918,0.0004361439,0.0001686962,0.0001831379],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004248829,0.0001686849,0.01388915,0.0006411644,0.00006200973,0.0004169153,0.003742004,0.01299338,0.8206933,0.00008416747,0.003322663,0.1435617],"study_design_scores_gemma":[0.0007587655,0.0009074617,0.001419806,0.00008925336,0.00002770186,0.00008896062,0.00002351375,0.2529406,0.5593077,0.002193686,0.1817867,0.0004557813],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7345099,0.0009090651,0.2435693,0.005125999,0.01046031,0.004113103,0.00001584286,0.001163006,0.0001334353],"genre_scores_gemma":[0.9943578,0.000006235121,0.004271616,0.0001751406,0.00009793064,0.0003513095,0.00001765778,0.00002490582,0.0006973686],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2613856,"threshold_uncertainty_score":0.8388357,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02582360075406547,"score_gpt":0.2599592462029758,"score_spread":0.2341356454489104,"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."}}