Exploring a dynamic digital biomarker of illness severity in older ICU patients via interpretable time-series deep learning modeling
Abstract Older adults admitted to intensive care units (ICUs) experience rapid physiological deterioration, pronounced heterogeneity, and limited physiological reserve, necessitating continuous and individualized assessment of illness severity. Dynamic digital biomarkers derived from high-frequency electronic health record data offer a promising solution for capturing evolving patient states in real time, yet their development and validation in elderly ICU populations remain limited, particularly with respect to interpretability and cross-institutional generalizability. In this international multicenter study, we developed an interpretable time-series deep learning model (Elder-DDB) to derive a dynamic digital biomarker of illness severity in critically ill older adults, operationalized as a continuously updated, model-derived mortality risk trajectory over time. Using longitudinal data from more than 200 hospitals across the United States, Europe, and Asia, the model was trained on 41,050 elderly ICU admissions and externally validated on 47,335 admissions from independent cohorts. Across all external validation cohorts, the model demonstrated robust and consistent discrimination (AUROC range: 0.801–0.834), significantly outperforming conventional clinical scores and state-of-the-art deep learning baselines. Interpretability analyses confirm the proposed dynamic digital biomarker reliably assesses real-time illness severity in elderly ICU patients, supporting precision critical care.
Authors
- Zhengbo Zhang (ORCID: https://orcid.org/0000-0001-9218-5644)
- Zhenyue Gao (ORCID: https://orcid.org/0000-0003-2283-2820)
- Sicheng Hao (ORCID: https://orcid.org/0000-0002-8905-005X)
- Minjie Duan
- Muyang Yan (ORCID: https://orcid.org/0000-0001-6978-1472)
- Yuhong Liu (ORCID: https://orcid.org/0000-0002-6478-2211)
- Chao Liu (ORCID: https://orcid.org/0000-0001-9257-4519)
- Lin Chen (ORCID: https://orcid.org/0000-0002-8720-9987)
- Pan Hu (ORCID: https://orcid.org/0009-0006-1796-6866)
- Wei Yan (ORCID: https://orcid.org/0000-0002-9495-6155)
- Yuqiang Wang (ORCID: https://orcid.org/0000-0002-0065-2004)
- Lijing Jia
- Zhongheng Zhang
- Xiaoli Liu
- Yue Zhang
- Hongjun Kang
Institutions
- National University of Singapore (SG)
- Shaoxing University (CN)
- Sichuan University (CN)
- Sir Run Run Shaw Hospital (CN)
- Chinese PLA General Hospital (CN)
- West China Hospital of Sichuan University (CN)
- People's Liberation Army 401 Hospital (CN)
- PLA 306 Hospital (CN)
- First Affiliated Hospital of Chinese PLA General Hospital (CN)
- Lishui Academy of Agricultural Sciences (CN)
- Second Affiliated Hospital of Zhejiang University (CN)
- PLA Academy of Military Science (CN)
- The Eighth Hospital of Xi'an (CN)
- Zhejiang University (CN)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2026-08-20
- DOI
- https://doi.org/10.1038/s41746-026-03114-2
- Primary Topic
- Sepsis Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China
- Zhejiang University
- National Science and Technology Major Project
- Wu Jieping Medical Foundation
- National Key Research and Development Program of China
- Zhejiang Traditional Chinese Medicine Administration
- Fundamental Research Funds for the Central Universities
- Natural Science Foundation of Zhejiang Province
- Science and Technology Program of Zhejiang Province