{"id":"W4281856165","doi":"10.1101/2022.05.24.22275490","title":"Evaluating the impact on clinical task efficiency of a natural language processing algorithm for searching medical documents: Prospective crossover study","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre for Global Health Research","funders":"University of Edinburgh; UK Research and Innovation","keywords":"Computer science; Task (project management); Workflow; Upload; Artificial intelligence; Natural language processing; Information retrieval; String searching algorithm; Search algorithm; Machine learning; Algorithm; World Wide Web; Pattern matching; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005256066,0.0002958599,0.0004710804,0.00005704555,0.0002790093,0.00007436064,0.0009250056,0.0002978533,0.00004856674],"category_scores_gemma":[0.004034784,0.0001717233,0.0003700316,0.000130722,0.0004047511,0.000002481227,0.00135226,0.001263108,9.489489e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005574265,"about_ca_system_score_gemma":0.0006873303,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007792412,"about_ca_topic_score_gemma":0.000008735037,"domain_scores_codex":[0.9965194,0.0006719421,0.0006312733,0.0008120677,0.0009746779,0.0003906427],"domain_scores_gemma":[0.998378,0.0003897046,0.0003989274,0.0005796539,0.0001480076,0.0001056433],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001487637,0.002178893,0.05707895,0.0001819931,0.0008243791,0.00003696957,0.004767799,0.0003253881,0.002643459,0.000003735083,0.0004166048,0.9300542],"study_design_scores_gemma":[0.02797434,0.09670116,0.4043287,0.001750783,0.001119003,0.0001139061,0.03182585,0.4203544,0.00867424,0.001015513,0.002470331,0.003671764],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9930854,0.002344485,0.002334028,0.0001376052,0.0005695999,0.001320136,0.00009682518,0.00002788902,0.00008401277],"genre_scores_gemma":[0.9967124,0.00002018701,0.001974319,0.0001015741,0.0005036583,0.0003311863,0.0001143248,0.00003290602,0.0002094689],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9263824,"threshold_uncertainty_score":0.7002672,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05282356704751386,"score_gpt":0.4958383169671636,"score_spread":0.4430147499196497,"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."}}