{"id":"W3178029650","doi":"10.1016/j.jbi.2021.103864","title":"Development of a generalizable natural language processing pipeline to extract physician-reported pain from clinical reports: Generated using publicly-available datasets and tested on institutional clinical reports for cancer patients with bone metastases","year":2021,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; McGill University Health Centre","funders":"","keywords":"Pipeline (software); 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":[0.004419673,0.0001914579,0.0006560604,0.0001842935,0.0001384062,0.00016714,0.0002048894,0.0001165566,0.00001019769],"category_scores_gemma":[0.002011565,0.0001364254,0.00009178145,0.0005306323,0.0001054783,0.0009245593,0.0001928721,0.0003235975,3.119147e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009688234,"about_ca_system_score_gemma":0.002483352,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002926416,"about_ca_topic_score_gemma":0.0000237806,"domain_scores_codex":[0.9943768,0.0001336514,0.003961449,0.0002803174,0.0009778352,0.0002698949],"domain_scores_gemma":[0.9951683,0.0002647399,0.002784353,0.0003878927,0.001056982,0.000337775],"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":[0.0002591773,0.002894422,0.01321387,0.0004101633,0.0008012924,0.001776189,0.001747091,0.003247078,0.002044423,0.00001543151,0.01752122,0.9560696],"study_design_scores_gemma":[0.002660078,0.0004401904,0.001774596,0.00129933,0.0002076409,0.0005683166,0.000285554,0.9367789,0.002741901,0.00003358092,0.05274799,0.0004619027],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5067648,0.0003740281,0.4920696,0.0001412191,0.0004214678,0.0001516496,0.00006121823,0.00001262668,0.000003351883],"genre_scores_gemma":[0.1355863,0.00001450623,0.8616301,0.001929107,0.0002669939,0.000005724756,0.0005454749,0.00001078426,0.00001101822],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9556078,"threshold_uncertainty_score":0.5563267,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07284276618582329,"score_gpt":0.3608962595140042,"score_spread":0.2880534933281809,"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."}}