{"id":"W6979281260","doi":"","title":"Reconstructing Sepsis Trajectories from Clinical Case Reports using LLMs: the Textual Time Series Corpus for Sepsis","year":2025,"lang":"en","type":"article","venue":"ArXiv.org","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. National Library of Medicine; Natural Sciences and Engineering Research Council of Canada; NHLBI Division of Intramural Research; Oak Ridge Institute for Science and Education; National Institutes of Health; Oak Ridge Associated Universities; Carnegie Mellon University; U.S. Department of Energy","keywords":"Timeline; Automatic summarization; Pipeline (software); Sepsis; Series (stratigraphy); Construct (python library); Time series","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":[0.001351359,0.0002547355,0.0004441517,0.00009110093,0.000935367,0.000212123,0.0005001417,0.0001920193,0.00004221678],"category_scores_gemma":[0.002569737,0.0002060953,0.0002244759,0.0005426703,0.0002312498,0.0004646558,0.0004066034,0.0005474833,0.00001523452],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009331673,"about_ca_system_score_gemma":0.0004100895,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002105217,"about_ca_topic_score_gemma":0.0002520879,"domain_scores_codex":[0.9970397,0.0004600583,0.001001439,0.0008455027,0.0001994699,0.0004538417],"domain_scores_gemma":[0.9953787,0.002566738,0.0005028418,0.001196789,0.0002411465,0.0001137288],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00001948278,0.00001981823,0.9815314,0.00003832183,0.00008316582,0.0006859679,0.0008382311,0.0001102552,0.00007326898,0.0005006121,0.0005257663,0.01557373],"study_design_scores_gemma":[0.002267115,0.0008918073,0.5514656,0.001167688,0.0006168659,0.01793511,0.006325191,0.3520156,0.00363894,0.01506643,0.04606915,0.002540553],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9732058,0.0002682867,0.01978959,0.003196543,0.002470955,0.0004449965,0.00002098859,0.0003101585,0.000292674],"genre_scores_gemma":[0.9434986,0.000009847914,0.05443574,0.0009169899,0.0004647425,0.00003701331,0.00001013137,0.00002444885,0.0006024677],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4300658,"threshold_uncertainty_score":0.8404324,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07219717221198486,"score_gpt":0.366233288556196,"score_spread":0.2940361163442111,"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."}}