{"id":"W4200369541","doi":"10.1109/embc46164.2021.9630347","title":"Analysis of Language Embeddings for Classification of Unstructured Pathology Reports","year":2021,"lang":"en","type":"article","venue":"2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)","topic":"AI in cancer detection","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Focus (optics); Digital pathology; Embedding; Word (group theory); Word embedding; Term (time)","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":[0.002286162,0.001437428,0.000594455,0.002555838,0.0003357759,0.001252549,0.0007122125,0.001040968,0.001583652],"category_scores_gemma":[0.010421,0.0001947925,0.000811611,0.001280555,0.0004670399,0.00188676,0.000922123,0.001280547,0.001420969],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000634938,"about_ca_system_score_gemma":0.0006859357,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002243815,"about_ca_topic_score_gemma":0.001622093,"domain_scores_codex":[0.9984053,0.0006421264,0.0001831046,0.0003088504,0.0002829452,0.0001778594],"domain_scores_gemma":[0.9913309,0.005750246,0.0006823704,0.0005292729,0.00146753,0.0002398148],"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":[0.001733792,0.001225991,0.03146403,0.0008710327,0.0003299322,0.0009839544,0.0009552857,0.1345034,0.02515401,0.004867508,0.02247336,0.7754378],"study_design_scores_gemma":[0.00003080403,0.0002230317,0.004887054,0.0000486244,0.00004715558,0.0002008183,0.0004741981,0.9826936,0.005812823,0.003485717,0.002060811,0.00003536944],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.750995,0.003031864,0.2319089,0.001366341,0.0006402948,0.0002559228,0.004870505,0.00385407,0.003077154],"genre_scores_gemma":[0.9271249,0.0005264369,0.05916072,0.0001116609,0.0001894826,0.0001506515,0.01013176,0.0001652131,0.002439318],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002555838,"threshold_uncertainty_score":0.01209056,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03128151688328491,"score_gpt":0.3206243122520457,"score_spread":0.2893427953687608,"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."}}