{"id":"W4396781237","doi":"10.1038/s43856-024-00502-1","title":"A comprehensive AI model development framework for consistent Gleason grading","year":2024,"lang":"en","type":"article","venue":"Communications Medicine","topic":"AI in cancer detection","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Agency for Science, Technology and Research","keywords":"Computer science; Workflow; Artificial intelligence; Scanner; Grading (engineering); Digital pathology; Scalability; Image quality; Pattern recognition (psychology); Machine learning; Computer vision; Image (mathematics); Database","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002853063,0.00136886,0.0007251225,0.001525321,0.0007186486,0.002027858,0.002954634,0.001289782,0.005227896],"category_scores_gemma":[0.007025807,0.0007072472,0.002009244,0.000820794,0.000611818,0.001464063,0.002101827,0.002638625,0.002314682],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001983897,"about_ca_system_score_gemma":0.004399942,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02341507,"about_ca_topic_score_gemma":0.03253348,"domain_scores_codex":[0.998966,0.0002456751,0.00008863776,0.0002299983,0.0003780037,0.00009178108],"domain_scores_gemma":[0.9977134,0.0009550154,0.0001441926,0.0002163287,0.000844115,0.000126937],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007539697,0.0001086847,0.002506827,0.000188192,0.0001356329,0.0001840181,0.0001849436,0.8176714,0.00350149,0.01820723,0.01069685,0.1465395],"study_design_scores_gemma":[0.000006038157,0.00001044505,0.00008917347,0.00001728311,0.00001345597,0.00002374441,0.00000905784,0.9895734,0.0006282692,0.00631149,0.003311133,0.000006558847],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003090605,0.0002428825,0.9891905,0.0004661455,0.00004679368,0.0001326993,0.000351625,0.004535024,0.001943746],"genre_scores_gemma":[0.1410558,0.000584138,0.8472564,0.00067292,0.0001226534,0.0007553961,0.0030167,0.0009997919,0.005536279],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02341507,"threshold_uncertainty_score":0.04655755,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.146219847374851,"score_gpt":0.3956946391391918,"score_spread":0.2494747917643409,"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."}}