{"id":"W4214578519","doi":"10.2196/35257","title":"Natural Language Processing for Assessing Quality Indicators in Free-Text Colonoscopy and Pathology Reports: Development and Usability Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Colorectal Cancer Screening and Detection","field":"Medicine","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Research Foundation of Korea; National Research Foundation","keywords":"Colonoscopy; Artificial intelligence; Pipeline (software); Computer science; Natural language processing; Medicine; Data set; Information extraction; Set (abstract data type); Medical physics; Colorectal cancer; Internal medicine; Cancer","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.05403955,0.00116192,0.000768579,0.00417189,0.0006125853,0.001788848,0.00165169,0.001061745,0.001101251],"category_scores_gemma":[0.120318,0.0006191765,0.001495519,0.002262249,0.0008343156,0.003164065,0.002087128,0.000854241,0.0006264317],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00114666,"about_ca_system_score_gemma":0.002207312,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002916368,"about_ca_topic_score_gemma":0.003140721,"domain_scores_codex":[0.9708335,0.01903295,0.00357911,0.002681164,0.003522502,0.0003507368],"domain_scores_gemma":[0.7361575,0.2219933,0.005740246,0.007565074,0.02750297,0.001040931],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002516828,0.003915621,0.08852693,0.007902116,0.0007341292,0.001986728,0.02911682,0.006165767,0.04224597,0.0008966733,0.008302273,0.8076902],"study_design_scores_gemma":[0.003613404,0.01627897,0.473898,0.004238035,0.002911101,0.008755285,0.02352187,0.289037,0.1092822,0.004275986,0.06301627,0.001171934],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.824551,0.001150925,0.1524896,0.0006186013,0.0001074264,0.01034805,0.002566443,0.005671077,0.002496893],"genre_scores_gemma":[0.5294978,0.0006779297,0.4583112,0.0003026611,0.00006644981,0.005794806,0.003826991,0.0005312377,0.000990966],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05403955,"threshold_uncertainty_score":0.2857919,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02101719244389338,"score_gpt":0.3721546888883932,"score_spread":0.3511374964444998,"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."}}