通过个体化寿命轨迹建模重新评估死亡风险
Re-timing mortality risk through individualized modelling of lifespan trajectories
该研究利用UK Biobank数据,通过个体化建模重新评估寿命轨迹中的死亡风险。研究数据来自UK Biobank和SEED研究,但个体级数据因伦理和隐私考虑无法公开共享。相关代码已发布在GitHub上。
Data availability
The UK Biobank data used in this study were accessed under approved application numbers 45925 and 86091. Owing to ethical and privacy considerations, individual-level data from the SEED study cannot be shared publicly. Researchers who meet the necessary institutional and ethical requirements may request access by contacting the Singapore Eye Research Institute (SERI) at seri@seri.com.sg. The individual-level UEMS data are available from the principal investigators of the Ural Eye and Medical Study upon reasonable request and with appropriate ethical approval. The Age-Related Eye Disease Study data are publicly available in the Database of Genotypes and Phenotypes and can be obtained on request from https://dbgap.ncbi.nlm.nih.gov/aa/wga.cgi?page=login. Source data are provided with this paper.
Code availability
The code used in the current study to develop the algorithm is available via GitHub at https://github.com/xt0402/RetiME.
References
Ageing and Health (World Health Organization, 2024) https://www.who.int/news-room/fact-sheets/detail/ageing-and-health
Woolf, S. H. A closer look at the economic argument for disease prevention. JAMA 301, 536–538 (2009).
Preventive Care: A National Profile on Use, Disparities, and Health Benefits (Partnership for Prevention, 2007) https://www.policyarchive.org/handle/10207/21574
Robinson, O. et al. Determinants of accelerated metabolomic and epigenetic aging in a UK cohort. Aging Cell 19, e13149 (2020).
Nie, C. et al. Distinct biological ages of organs and systems identified from a multi-omics study. Cell Rep. 38, 110459 (2022).
Lu, A. T. et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging 11, 303–327 (2019).
Levine, M. E. et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging 10, 573–591 (2018).
Hannum, G. et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol. Cell 49, 359–367 (2013).
Nakamura, E. & Miyao, K. A method for identifying biomarkers of aging and constructing an index of biological age in humans. J. Gerontol. A Biol. Sci. Med. Sci. 62, 1096–1105 (2007).
Tian, Y. E. et al. Heterogeneous aging across multiple organ systems and prediction of chronic disease and mortality. Nat. Med. 29, 1221–1231 (2023).
Drewelies, J. et al. Using blood test parameters to define biological age among older adults: association with morbidity and mortality independent of chronological age validated in two separate birth cohorts. GeroScience 44, 2685–2699 (2022).
Fong, S. et al. Principal component-based clinical aging clocks identify signatures of healthy aging and targets for clinical intervention. Nat. Aging 4, 1137–1152 (2024).
Horvath, S. DNA methylation age of human tissues and cell types. Genome Biol. 14, 1–20 (2013).
Klemera, P. & Doubal, S. A new approach to the concept and computation of biological age. Mech. Ageing Dev. 127, 240–248 (2006).
Pyrkov, T. V. et al. Extracting biological age from biomedical data via deep learning: too much of a good thing? Sci. Rep. 8, 5210 (2018).
Ikram, M. K. et al. Retinal vascular caliber as a biomarker for diabetes microvascular complications. Diabetes Care 36, 750–759 (2013).
McGeechan, K. et al. Meta-analysis: retinal vessel caliber and risk for coronary heart disease. Ann. Intern. Med. 151, 404–413 (2009).
Hinton, D. R., Sadun, A. A., Blanks, J. C. & Miller, C. A. Optic-nerve degeneration in Alzheimer’s disease. N. Engl. J. Med. 315, 485–487 (1986).
Nusinovici, S. et al. Retinal photograph-based deep learning predicts biological age, and stratifies morbidity and mortality risk. Age Ageing 51, afac065 (2022).
Zhu, Z. et al. Retinal age gap as a predictive biomarker for mortality risk. Br. J. Ophthalmol. 107, 547–554 (2023).
Nusinovici, S. et al. Application of a deep-learning marker for morbidity and mortality prediction derived from retinal photographs: a cohort development and validation study. Lancet Healthy Longev. 5, 100593 (2024).
Quan, H. et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am. J. Epidemiol. 173, 676–682 (2011).
Quan, H. et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med. Care 43, 1130–1139 (2005).
Charlson, M. E., Carrozzino, D., Guidi, J. & Patierno, C. Charlson comorbidity index: a critical review of clinimetric properties. Psychother. Psychosom. 91, 8–35 (2022).
Sudlow, C. et al. UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 12, e1001779 (2015).
Majithia, S. et al. Corrigendum to: Cohort profile: the Singapore Epidemiology of Eye Diseases Study (SEED). Int. J. Epidemiol. 50, 1401 (2021).
Bikbov, M. M. et al. Mortality and ocular parameters and diseases. Eye 39, 1322–1331 (2025).
Age-Related Eye Disease Study Research Group The Age-Related Eye Disease Study (AREDS): design implications. AREDS report no. 1. Control. Clin. Trials 20, 573–600 (1999).
UK Biobank Death Summary Report (UK Biobank, 2025) https://biobank.ndph.ox.ac.uk/~bbdatan/DeathSummaryReport.html
Leading Causes of Death, UK: 2001 to 2018 (Office for National Statistics, 2020) https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/causesofdeath/articles/leadingcausesofdeathuk/2001to2018
Principal Causes of Death (2024) [Dataset] (Ministry of Health, 2024) https://data.gov.sg/datasets/d_48143a2b16027afcadeb362352b0266a/view
Wilson, P. W. et al. Prediction of coronary heart disease using risk factor categories. Circulation 97, 1837–1847 (1998).
Goff, D. C. Jr. et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J. Am. Coll. Cardiol. 63, 2935–2959 (2014).
Austin, P. C., Lee, D. S. & Fine, J. P. Introduction to the analysis of survival data in the presence of competing risks. Circulation 133, 601–609 (2016).
Wang, Z. et al. A general model for retinal segmentation and quantification. Preprint at https://doi.org/10.48550/arXiv.2602.07012 (2026).
Gompertz, B. XXIV. On the nature of the function expressive of the law of human mortality, and on a new mode of determining the value of life contingencies. In a letter to Francis Baily, Esq. FRS &c. Phil. Trans. R. Soc. Lond. 115, 513–583 (1825).
Makeham, W. M. On the law of mortality and the construction of annuity tables. Assur. Mag. J. Inst. Actuar. 8, 301–310 (1860).
Missov, T. I. & Lenart, A. Gompertz–Makeham life expectancies: expressions and applications. Theor. Popul. Biol. 90, 29–35 (2013).
Population, United Kingdom of Great Britain and Northern Ireland (World Health Organization, 2023) https://data.who.int/countries/826
Wang, M. et al. Enhancing diagnostic accuracy in rare and common fundus diseases with a knowledge-rich vision-language model. Nat. Commun. 16, 5528 (2025).
Hu, E. J. et al. LoRA: low-rank adaptation of large language models. ICLR 1, 3 (2022).
de Leeuw, C. A., Mooij, J. M., Heskes, T. & Posthuma, D. MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput. Biol. 11, e1004219 (2015).
Acknowledgements
We thank the participants and study teams of the UK Biobank, SEED, UEMS and AREDS cohorts.
Funding
This research is supported by the Singapore Ministry of Health’s National Medical Research Council under its Clinician Scientist—Individual Research Grant (MOH-001477-00), and its Healthy and Meaningful Longevity (HML)—Cognition Grant (MOH-001876-00; Retina-Enhanced Cognitive-impairment Opportunistic and Generative Neural Intelligent Screening Evaluation, RECOGNISE), and by the NUS-NUHS Ageing and Longevity Grant (FY25 AHP/10/Ageing&Longevity-Feb25/07), National University Health System, Singapore. C.-Y.C. is supported by the Singapore Ministry of Health’s National Medical Research Council under its Senior Clinician Scientist Award (MOH-001283-00). D.L. is supported by the Ministry of Education Singapore (MOE/T2EP20125-0026). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Author information
Author notes
These authors contributed equally: Ting Xu, Qingsheng Peng.
These authors jointly supervised this work: Dianbo Liu, Ching-Yu Cheng.
Authors and Affiliations
Center for Innovation and Precision Eye Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore
Ting Xu, Hengtong Li, Meng Wang, Hongyu He, Dianbo Liu & Ching-Yu Cheng
Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore
Ting Xu, Hengtong Li, Meng Wang, Hongyu He, Dianbo Liu & Ching-Yu Cheng
Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore
Qingsheng Peng, Marco Yu, Crystal Chong, Can Can Xue, Xiayin Zhang, Jost B. Jonas & Ching-Yu Cheng
Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore, Singapore
Qingsheng Peng & Ching-Yu Cheng
Center for Frontier AI Research (CFAR), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore
Jie Zhang
National Eye Institute, National Institutes of Health, Bethesda, MD, USA
Emily Y. Chew
Ufa Eye Research Institute, Bashkir State Medical University, Ufa, Russia
Mukharram M. Bikbov & Gyulli M. Kazakbaeva
Rothschild Foundation Hospital, Paris, France
Jost B. Jonas
L V Prasad Eye Institute, Hyderabad, India
Jost B. Jonas
Beijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital, Tsinghua Medicine, Tsinghua University, Beijing, China
Jost B. Jonas
Institute of Advanced Intelligence and Computing (IAIC), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore
Huazhu Fu
Authors
- Ting Xu
- Qingsheng Peng
- Hengtong Li
- Meng Wang
- Jie Zhang
- Marco Yu
- Crystal Chong
- Can Can Xue
- Xiayin Zhang
- Hongyu He
- Emily Y. Chew
- Mukharram M. Bikbov
- Gyulli M. Kazakbaeva
- Jost B. Jonas
- Huazhu Fu
- Dianbo Liu
- Ching-Yu Cheng
Contributions
T.X.: conceptualization, methodology, AI model development, data analysis, writing (original draft), visualization and project coordination. Q.P.: clinical conceptualization, clinical data interpretation, validation, writing (original draft) and supervision. H.L.: genomic data analysis, integration of genetic and imaging features, writing, review and editing. M.W.: AI model implementation, experimental design, writing, review and editing. J.Z.: model optimization, ablation analysis, writing, review and editing. M.Y.: data preprocessing and statistical analysis. C.C.: external validation and independent evaluation. C.C.X.: clinical assessment and phenotypic data curation. X.Z.: clinical guidance and paper review. H.H.: model evaluation. H.F.: technical guidance on AI model architecture and integration, writing, review and editing. E.Y.C.: provision of the AREDS dataset, interpretation of AREDS validation results and paper review. M.M.B., G.M.K. and J.B.J.: provision of the UEMS dataset, interpretation of UEMS validation results and paper review. D.L.: project supervision, AI–clinical translation strategy, methodology, writing, review and editing. C.-Y.C.: overall supervision, clinical oversight, cohort governance, methodology, writing, review and editing. All authors reviewed and approved the final paper.
Corresponding author
Correspondence to Ching-Yu Cheng.
Ethics declarations
Competing interests
T.X., Q.P., D.L. and C.-Y.C. are named as inventors on a provisional patent application (application number 10202601434W) filed by the National University of Singapore, covering the retinal image-based method for individualized mortality-risk estimation (RetiME) described in this paper. C.-Y.C. is the co-founder of Eye.AI and a consultant of MediWhale. The other authors declare no competing interests.
Peer review
Peer review information
Nature Biomedical Engineering thanks Carol Cheung, Yalin Wang and Zhuoting Zhu for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Source data
Source Data Fig. 2 (download XLSX )
Statistical source data: hazard ratios, 95% confidence intervals and FDR-adjusted P values for all disease outcomes and ageing markers plotted in Fig. 2.
Source Data Fig. 3 (download XLSX )
Statistical source data for panel d (kernel density curves of observed age at death and RetiME-estimated T-max). Participant-level UK Biobank data underlying panels a–c and e are not provided, as they cannot be redistributed under the UK Biobank Material Transfer Agreement; they are available to approved researchers via application to UK Biobank.
Source Data Fig. 4 (download CSV )
Statistical source data: cause-specific hazard ratios with exact P values underlying the heat map in Fig. 4c.
Source Data Fig. 5 (download XLSX )
Statistical source data for panel a (mean absolute SHAP values per feature and per biological domain). Participant-level UK Biobank data underlying the SHAP dependence plots in panel b are not provided, as they cannot be redistributed under the UK Biobank Material Transfer Agreement; they are available to approved researchers via application to UK Biobank.
Source Data Fig. 6 (download XLSX )
Statistical source data: protein-level regression estimates (beta, raw and FDR-adjusted P values, panel assignment) and pathway/disease enrichment statistics plotted in Fig. 6.
About this article
Cite this article
Xu, T., Peng, Q., Li, H. et al. Re-timing mortality risk through individualized modelling of lifespan trajectories. Nat. Biomed. Eng (2026). https://doi.org/10.1038/s41551-026-01822-4
Received: 13 November 2025
Accepted: 25 September 2026
Published: 02 November 2026
Version of record: 02 November 2026
DOI: https://doi.org/10.1038/s41551-026-01822-4
来源:Nature 健康护理主题 · nature.com