{"id":"W4415158621","doi":"10.1007/s13755-025-00384-0","title":"Timelygpt: extrapolatable transformer pre-training for long-term time-series forecasting in healthcare","year":2025,"lang":"en","type":"article","venue":"Health Information Science and Systems","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Health care; Trajectory; State (computer science); Key (lock); Health data","routes":{"ca_aff":true,"ca_fund":true,"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":[0.005798754,0.000163088,0.0003414949,0.000801997,0.0009971885,0.0006242309,0.0006092602,0.00008169782,0.00000155194],"category_scores_gemma":[0.0003680658,0.0001506964,0.00002821093,0.001750717,0.0001133669,0.006267549,0.00006919855,0.0002320393,0.000007905974],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003159359,"about_ca_system_score_gemma":0.002272691,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007081057,"about_ca_topic_score_gemma":0.0001099903,"domain_scores_codex":[0.9971448,0.0001049687,0.001040673,0.0003385805,0.0005672289,0.0008038045],"domain_scores_gemma":[0.998448,0.0001823222,0.0003305259,0.0003545921,0.0004518846,0.0002327158],"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":[0.00008521772,0.00002852364,0.103,0.01134123,0.00000945182,0.000001876417,0.05163929,0.001954412,0.00004884996,0.1900639,0.0007519228,0.6410753],"study_design_scores_gemma":[0.0007644314,0.0003425288,0.06861153,0.00124418,0.000001679343,0.00007466933,0.0007129918,0.9186961,0.00003426115,0.0003820643,0.008849621,0.0002859508],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07241926,0.002607352,0.8646021,0.04612596,0.002686285,0.006511737,0.00005351094,0.0005305373,0.004463261],"genre_scores_gemma":[0.98997,0.00007134908,0.00672229,0.002603533,0.00004421032,0.0002742951,0.00002041723,0.000006402578,0.000287562],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9175507,"threshold_uncertainty_score":0.7669669,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06041971627833059,"score_gpt":0.3547379458841999,"score_spread":0.2943182296058693,"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."}}