跳到正文
原文
Nature 健康护理主题·· 18 天前AI 评分35

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

AI 导读

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。

正文

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

  1. Ageing and Health (World Health Organization, 2024) https://www.who.int/news-room/fact-sheets/detail/ageing-and-health

  2. Woolf, S. H. A closer look at the economic argument for disease prevention. JAMA 301, 536–538 (2009).

    Article  CAS  PubMed  Google Scholar 

  3. Preventive Care: A National Profile on Use, Disparities, and Health Benefits (Partnership for Prevention, 2007) https://www.policyarchive.org/handle/10207/21574

  4. Robinson, O. et al. Determinants of accelerated metabolomic and epigenetic aging in a UK cohort. Aging Cell 19, e13149 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Nie, C. et al. Distinct biological ages of organs and systems identified from a multi-omics study. Cell Rep. 38, 110459 (2022).

    Article  CAS  PubMed  Google Scholar 

  6. Lu, A. T. et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging 11, 303–327 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Levine, M. E. et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging 10, 573–591 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  8. Hannum, G. et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol. Cell 49, 359–367 (2013).

    Article  CAS  PubMed  Google Scholar 

  9. 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).

    Article  PubMed  Google Scholar 

  10. Tian, Y. E. et al. Heterogeneous aging across multiple organ systems and prediction of chronic disease and mortality. Nat. Med. 29, 1221–1231 (2023).

    Article  CAS  PubMed  Google Scholar 

  11. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  12. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  13. Horvath, S. DNA methylation age of human tissues and cell types. Genome Biol. 14, 1–20 (2013).

    Article  Google Scholar 

  14. Klemera, P. & Doubal, S. A new approach to the concept and computation of biological age. Mech. Ageing Dev. 127, 240–248 (2006).

    Article  PubMed  Google Scholar 

  15. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  16. Ikram, M. K. et al. Retinal vascular caliber as a biomarker for diabetes microvascular complications. Diabetes Care 36, 750–759 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  17. McGeechan, K. et al. Meta-analysis: retinal vessel caliber and risk for coronary heart disease. Ann. Intern. Med. 151, 404–413 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  18. 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).

    Article  CAS  PubMed  Google Scholar 

  19. Nusinovici, S. et al. Retinal photograph-based deep learning predicts biological age, and stratifies morbidity and mortality risk. Age Ageing 51, afac065 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

  20. Zhu, Z. et al. Retinal age gap as a predictive biomarker for mortality risk. Br. J. Ophthalmol. 107, 547–554 (2023).

    Article  PubMed  Google Scholar 

  21. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  22. 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).

    Article  PubMed  Google Scholar 

  23. Quan, H. et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med. Care 43, 1130–1139 (2005).

    Article  PubMed  Google Scholar 

  24. Charlson, M. E., Carrozzino, D., Guidi, J. & Patierno, C. Charlson comorbidity index: a critical review of clinimetric properties. Psychother. Psychosom. 91, 8–35 (2022).

    Article  PubMed  Google Scholar 

  25. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  26. Majithia, S. et al. Corrigendum to: Cohort profile: the Singapore Epidemiology of Eye Diseases Study (SEED). Int. J. Epidemiol. 50, 1401 (2021).

    Article  PubMed  Google Scholar 

  27. Bikbov, M. M. et al. Mortality and ocular parameters and diseases. Eye 39, 1322–1331 (2025).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. 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).

    Article  Google Scholar 

  29. UK Biobank Death Summary Report (UK Biobank, 2025) https://biobank.ndph.ox.ac.uk/~bbdatan/DeathSummaryReport.html

  30. Leading Causes of Death, UK: 2001 to 2018 (Office for National Statistics, 2020) https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/causesofdeath/articles/leadingcausesofdeathuk/2001to2018

  31. Principal Causes of Death (2024) [Dataset] (Ministry of Health, 2024) https://data.gov.sg/datasets/d_48143a2b16027afcadeb362352b0266a/view

  32. Wilson, P. W. et al. Prediction of coronary heart disease using risk factor categories. Circulation 97, 1837–1847 (1998).

    Article  CAS  PubMed  Google Scholar 

  33. 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).

    Article  PubMed  Google Scholar 

  34. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  35. Wang, Z. et al. A general model for retinal segmentation and quantification. Preprint at https://doi.org/10.48550/arXiv.2602.07012 (2026).

  36. 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).

  37. Makeham, W. M. On the law of mortality and the construction of annuity tables. Assur. Mag. J. Inst. Actuar. 8, 301–310 (1860).

    Article  Google Scholar 

  38. Missov, T. I. & Lenart, A. Gompertz–Makeham life expectancies: expressions and applications. Theor. Popul. Biol. 90, 29–35 (2013).

    Article  PubMed  Google Scholar 

  39. Population, United Kingdom of Great Britain and Northern Ireland (World Health Organization, 2023) https://data.who.int/countries/826

  40. 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).

    Article  PubMed  PubMed Central  Google Scholar 

  41. Hu, E. J. et al. LoRA: low-rank adaptation of large language models. ICLR 1, 3 (2022).

    Google Scholar 

  42. 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).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

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

  1. These authors contributed equally: Ting Xu, Qingsheng Peng.

  2. These authors jointly supervised this work: Dianbo Liu, Ching-Yu Cheng.

Authors and Affiliations

  1. 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

  2. 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

  3. 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

  4. Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore, Singapore

    Qingsheng Peng & Ching-Yu Cheng

  5. Center for Frontier AI Research (CFAR), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore

    Jie Zhang

  6. National Eye Institute, National Institutes of Health, Bethesda, MD, USA

    Emily Y. Chew

  7. Ufa Eye Research Institute, Bashkir State Medical University, Ufa, Russia

    Mukharram M. Bikbov & Gyulli M. Kazakbaeva

  8. Rothschild Foundation Hospital, Paris, France

    Jost B. Jonas

  9. L V Prasad Eye Institute, Hyderabad, India

    Jost B. Jonas

  10. Beijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital, Tsinghua Medicine, Tsinghua University, Beijing, China

    Jost B. Jonas

  11. Institute of Advanced Intelligence and Computing (IAIC), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore

    Huazhu Fu

Authors

  1. Ting Xu
  2. Qingsheng Peng
  3. Hengtong Li
  4. Meng Wang
  5. Jie Zhang
  6. Marco Yu
  7. Crystal Chong
  8. Can Can Xue
  9. Xiayin Zhang
  10. Hongyu He
  11. Emily Y. Chew
  12. Mukharram M. Bikbov
  13. Gyulli M. Kazakbaeva
  14. Jost B. Jonas
  15. Huazhu Fu
  16. Dianbo Liu
  17. 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

Extended Data Table 1 Associations between RetiME and the top 10 cause-specific mortality risks in the UK

Full size table

Extended Data Table 2 Associations between RetiME and the top 20 cause-specific mortalities in the UK Biobank using Fine–Gray sub-distribution hazard models

Full size table

Extended Data Table 3 Associations between RetiME and the top 10 cause-specific mortalities in the UK using Fine–Gray sub-distribution hazard models

Full size table

Extended Data Table 4 Associations between RetiME and the top 10 cause-specific mortalities in Singapore in the external SEED Cohort

Full size table

Extended Data Table 5 Associations between RetiME and the risk of cause-specific mortality in the UEMS external dataset

Full size table

Extended Data Table 6 Associations between RetiME and cause-specific mortalities in the external UEMS cohort using Fine–Gray sub-distribution hazard models

Full size table

Extended Data Table 7 Lead SNPs reaching genome-wide significance levels

Full size table

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

Check for updates. Verify currency and authenticity via CrossMark

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

Download citation

  • 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