{"id":"W4405621264","doi":"10.48550/arxiv.2408.11861","title":"Speaking the Same Language: Leveraging LLMs in Standardizing Clinical Data for AI","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Health and Medical Research Council; Canadian Institutes of Health Research; Genentech; National Institutes of Health; IXICO; H. Lundbeck A/S; National Cancer Institute; Servier; Vetenskapsrådet; Eisai; Lunds Universitet; University of Melbourne; Royal Swedish Academy of Sciences; Commonwealth Scientific and Industrial Research Organisation; Northern California Institute for Research and Education; State Government of Victoria; BioClinica; Biogen; Pfizer; Novartis Pharmaceuticals Corporation; University of Southern California; Dementia Collaborative Research Centres, Australia; Edith Cowan University; Bristol-Myers Squibb; Eli Lilly and Company; Torsten Söderbergs Stiftelse; Alzheimer's Disease Neuroimaging Initiative; Medical Research Council; Meso Scale Diagnostics; Science and Industry Endowment Fund; Alzheimer's Association","keywords":"Computer science; Data science","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004515135,0.0002117481,0.0003452641,0.0001437956,0.0004758349,0.000315043,0.002399417,0.0003365091,0.0001033981],"category_scores_gemma":[0.001192081,0.0002019225,0.0002172516,0.0005156531,0.0006880265,0.0002770243,0.002999951,0.001346098,0.00007830544],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004193065,"about_ca_system_score_gemma":0.0007890608,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01309039,"about_ca_topic_score_gemma":0.03966427,"domain_scores_codex":[0.9973322,0.000494223,0.0004000897,0.001101839,0.0001737473,0.0004979424],"domain_scores_gemma":[0.9972073,0.001139985,0.0001897681,0.001225993,0.0001293792,0.0001075263],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002601878,0.0001865052,0.01954352,0.0003092254,0.0004008722,0.0009812865,0.06428617,0.07816123,0.00001883253,0.7823358,0.009782829,0.04373357],"study_design_scores_gemma":[0.0003496308,0.0000504021,0.0003617367,0.001100312,0.0004583978,0.000001504025,0.1219779,0.2733769,0.00005842837,0.4320089,0.1690954,0.001160523],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8754331,0.0009746057,0.08686815,0.005831425,0.006824976,0.002117698,0.0005036877,0.0004572735,0.02098909],"genre_scores_gemma":[0.9970115,0.0002722716,0.0002579091,0.0003393071,0.0009506984,0.000001530873,0.00003653944,0.00003006418,0.001100116],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3503269,"threshold_uncertainty_score":0.9934815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3928649249160623,"score_gpt":0.3865061568719257,"score_spread":0.006358768044136598,"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."}}