{"id":"W4308402150","doi":"10.5858/arpa.2022-0051-oa","title":"Overcoming the Interobserver Variability in Lung Adenocarcinoma Subtyping: A Clustering Approach to Establish a Ground Truth for Downstream Applications","year":2022,"lang":"en","type":"article","venue":"Archives of Pathology & Laboratory Medicine","topic":"Lung Cancer Diagnosis and Treatment","field":"Medicine","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vancouver General Hospital","funders":"","keywords":"Adenocarcinoma; Subtyping; Ground truth; Context (archaeology); Lung; Medicine; Pathology; Cluster analysis; Cluster (spacecraft); Radiology; Computer science; Internal medicine; Artificial intelligence; Biology; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02172605,0.001143322,0.001222627,0.003985526,0.001675317,0.003022222,0.002037098,0.002162507,0.001494194],"category_scores_gemma":[0.04649985,0.0007478221,0.001522156,0.002497214,0.001422402,0.001905913,0.002615584,0.001427729,0.0008486248],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00173402,"about_ca_system_score_gemma":0.001984891,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004610051,"about_ca_topic_score_gemma":0.006473249,"domain_scores_codex":[0.985662,0.007878751,0.0008906012,0.002923725,0.002221281,0.0004234257],"domain_scores_gemma":[0.9753174,0.0119793,0.002849248,0.00399475,0.00544297,0.0004164994],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001010054,0.0005154521,0.1320844,0.001329526,0.001606136,0.0004429103,0.005751312,0.1222779,0.06884052,0.009128115,0.01013746,0.6468762],"study_design_scores_gemma":[0.00008701955,0.0004067057,0.0910184,0.0002636845,0.0004059272,0.0004349683,0.001445112,0.8351465,0.03367288,0.02987102,0.007050625,0.0001971427],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09904907,0.000692402,0.8957329,0.0006741203,0.00009032644,0.0004557328,0.0003290218,0.001491815,0.001484526],"genre_scores_gemma":[0.4669549,0.0001682787,0.5310351,0.0001412384,0.00007310866,0.0003566684,0.0006199231,0.0002544915,0.0003963381],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02172605,"threshold_uncertainty_score":0.1148998,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0200681946436232,"score_gpt":0.2859516001975973,"score_spread":0.2658834055539741,"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."}}