{"id":"W4205601010","doi":"10.1002/path.5864","title":"Computer‐extracted features of nuclear morphology in hematoxylin and eosin images distinguish <scp>s</scp>tage <scp>II</scp> and <scp>IV</scp> colon tumors","year":2022,"lang":"en","type":"article","venue":"The Journal of Pathology","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"National Center for Advancing Translational Sciences; National Center for Research Resources; National Institute of Biomedical Imaging and Bioengineering; National Cancer Institute; National Heart, Lung, and Blood Institute; U.S. Department of Veterans Affairs","keywords":"Colorectal cancer; H&E stain; Artificial intelligence; Stage (stratigraphy); Digital pathology; Medicine; Random forest; Convolutional neural network; Cancer; Pathology; Oncology; Internal medicine; Computer science; Immunohistochemistry; Biology","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":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.002981699,0.0003084462,0.001105908,0.0004702067,0.000356886,0.00003094784,0.0004160582,0.0001746115,0.00001090055],"category_scores_gemma":[0.00562584,0.0002375938,0.0001322176,0.0003437502,0.0008598681,0.00008615205,0.0005531447,0.002457694,0.000001607054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007160954,"about_ca_system_score_gemma":0.0001141618,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006947344,"about_ca_topic_score_gemma":0.000005101628,"domain_scores_codex":[0.99674,0.0009919029,0.0009333708,0.0003321005,0.0004431844,0.000559413],"domain_scores_gemma":[0.9951561,0.003050413,0.001068473,0.0003055689,0.0001606187,0.000258844],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0002913218,0.002208979,0.1626025,0.000891648,0.0005729246,0.0360606,0.05592553,0.001825681,0.5995544,0.001786564,0.1264255,0.01185433],"study_design_scores_gemma":[0.006819917,0.004503528,0.8484898,0.0004162976,0.0004884029,0.1063641,0.009172613,0.005209296,0.002379236,0.001471592,0.01459392,0.00009124498],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9931791,0.003638881,0.0002536738,0.001512583,0.000474644,0.0002765605,0.00002384017,0.00003263933,0.0006080208],"genre_scores_gemma":[0.9938521,0.0005583507,0.003199096,0.001609276,0.0003309568,0.000004783391,0.00001349072,0.0000638481,0.0003680686],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6858873,"threshold_uncertainty_score":0.9998437,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007140593702023906,"score_gpt":0.2551362489681132,"score_spread":0.2479956552660893,"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."}}