{"id":"W3045125794","doi":"10.4103/jpi.jpi_10_20","title":"Deep Learning to Estimate Human Epidermal Growth Factor Receptor 2 Status from Hematoxylin and Eosin-Stained Breast Tissue Images","year":2020,"lang":"en","type":"article","venue":"Journal of Pathology Informatics","topic":"AI in cancer detection","field":"Computer Science","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"H&E stain; Immunohistochemistry; Breast cancer; Stain; Digital pathology; Pathology; Medicine; Human Epidermal Growth Factor Receptor 2; Artificial intelligence; Computer science; Staining; Internal medicine; 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":[0.0002951901,0.0001773834,0.0003781099,0.0001599137,0.0001700716,0.0001671664,0.0004217926,0.0001057836,0.00004011682],"category_scores_gemma":[0.0003086078,0.0001587792,0.00006277156,0.0002250502,0.00006617649,0.001215279,0.0002741491,0.000523394,0.00003984157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009114812,"about_ca_system_score_gemma":0.00006601245,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006976165,"about_ca_topic_score_gemma":0.000002457171,"domain_scores_codex":[0.9983004,0.0001161754,0.0008586101,0.0001207578,0.000283004,0.0003210403],"domain_scores_gemma":[0.9983093,0.0001437719,0.0007824757,0.0001503632,0.000273987,0.0003401694],"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":[0.0003686911,0.0001809231,0.04256085,0.0007722399,0.0004231319,0.001536941,0.2888531,0.01663433,0.1000819,0.002040426,0.00642854,0.5401189],"study_design_scores_gemma":[0.01279298,0.01607328,0.3303788,0.0007981651,0.0004261229,0.01924858,0.01755309,0.3964224,0.1719236,0.009457772,0.02087269,0.004052556],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3597958,0.00009044047,0.6377218,0.001879221,0.0002755535,0.00008902171,0.00001283902,0.00005635213,0.00007888823],"genre_scores_gemma":[0.7358937,0.00004446211,0.262978,0.0008524662,0.0002033233,0.000002365957,0.000003723629,0.00001382944,0.000008149103],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5360664,"threshold_uncertainty_score":0.6474826,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0131449571087564,"score_gpt":0.2807509058362703,"score_spread":0.2676059487275139,"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."}}