{"id":"W4289130391","doi":"10.1007/s00234-022-03029-1","title":"Rule-based natural language processing for automation of stroke data extraction: a validation study","year":2022,"lang":"en","type":"article","venue":"Neuroradiology","topic":"Acute Ischemic Stroke Management","field":"Medicine","cited_by":18,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto; Sunnybrook Health Science Centre; University of Ottawa; Health Sciences Centre; Ottawa Hospital","funders":"","keywords":"Medicine; Neuroradiology; Occlusion; Collateral circulation; Stroke (engine); Radiology; Angiography; Computed tomography angiography; Data extraction; Artificial intelligence; Natural language processing; Neurology; MEDLINE; Internal medicine; Computer science","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":[0.0003210477,0.00008723408,0.0002118077,0.0001407703,0.00009108728,0.000005883193,0.0002009333,0.00002346338,0.00008165542],"category_scores_gemma":[0.0001809247,0.00008592987,0.00003382958,0.0001371072,0.0000262484,0.00008313693,0.00013721,0.0001725223,0.000001517247],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006396083,"about_ca_system_score_gemma":0.00006556272,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009032799,"about_ca_topic_score_gemma":9.749239e-7,"domain_scores_codex":[0.9989986,0.0001098069,0.0002469356,0.0003093058,0.000204911,0.0001304121],"domain_scores_gemma":[0.9991525,0.0001115251,0.0002023676,0.0004620643,0.00004918426,0.00002239298],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003915566,0.003528139,0.1059977,0.001114973,0.0005064741,0.0002766601,0.002978545,0.002678037,0.6165727,0.00005407687,0.06921884,0.1931583],"study_design_scores_gemma":[0.02212534,0.008153068,0.2174691,0.00004995917,0.001830981,0.000672799,0.01234121,0.6578877,0.02939479,0.00001187925,0.0494441,0.0006191267],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9920587,0.0001497171,0.004775701,0.0009090506,0.0003718653,0.001181599,0.0001000297,0.00008399281,0.0003693303],"genre_scores_gemma":[0.9951079,5.385269e-7,0.002932407,0.0003389155,0.0001143154,0.0001555329,0.0009094941,0.00001576603,0.000425119],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6552097,"threshold_uncertainty_score":0.3504118,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03996540448498499,"score_gpt":0.3494512144862339,"score_spread":0.3094858100012489,"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."}}