Timelygpt: extrapolatable transformer pre-training for long-term time-series forecasting in healthcare
Together, we envision TimelyGPT to be useful in various health domains, including long-term patient health state forecasting, patient risk trajectory prediction, and disease classification. Its code is available at Github.
Authors
- Qincheng Lu
- David L. Buckeridge (ORCID: https://orcid.org/0000-0003-1817-5047)
- Yue Li (ORCID: https://orcid.org/0000-0003-3844-4865)
- He Zhu (ORCID: https://orcid.org/0009-0002-7917-4857)
- Ziyang Song (ORCID: https://orcid.org/0009-0009-6348-8713)
- Hao Xu (ORCID: https://orcid.org/0009-0009-6706-3704)
- Ziyang Song
Institutions
- Ohio University (US)
- Mila - Quebec Artificial Intelligence Institute (CA)
- McGill University (CA)
Publication Details
- Journal
- Health Information Science and Systems
- Published
- 2025-10-14
- DOI
- https://doi.org/10.1007/s13755-025-00384-0
- Primary Topic
- Time Series Analysis and Forecasting
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Natural Sciences and Engineering Research Council of Canada
- Fonds de recherche du Québec – Nature et technologies