跳到正文
原文
Nature 健康护理主题·· 25 天前AI 评分55

生物年龄与人工智能在医疗护理中的应用

The use of biological age and artificial intelligence in health care

AI 导读

本文强调生物年龄和AI在医疗中的应用,提出通过整合生物年龄和AI技术,可以提高临床精准度。文章讨论了生物年龄评估方法和AI在健康护理中的作用,强调了技术进步对精准医疗的影响。

正文

This commentary highlights the age illusion and advocates integrating biological age and artificial intelligence (AI) to enhance clinical precision in health care. By synthesizing geroscience and machine learning, practitioners and scientists can effectively bridge the gap between experimental biomarkers and personalized medicine and longevity care. This contribution is timely as the importance of biological age and predictive analytics offer transformative potential for diagnostics while necessitating a focus on modern medical systems. This article aims to elucidate how AI accelerates the clinical implementation of biological age rather than current chronological age in precision medicine.

Biological age estimations

The clinical overreliance on chronological age often fails to capture the significant physiological variability among individuals and requires further elucidation1. Importantly, emerging causality-informed epigenetic clocks, such as DamAge, distinguish accumulated molecular damage from adaptive responses, offering superior predictive power for morbidity compared with chronological metrics2,3. Furthermore, recent plasma protein-based models demonstrate that organs age at distinct rates, providing a granular view of system-specific vulnerabilities to diseases such as heart failure and dementia4,5. To integrate these findings into clinical practice, we propose implementing multi-omic frameworks that integrate proteomics and DNA methylation within standardized platforms like Biolearn to guide high-stakes decisions, such as surgery eligibility and intensive care triage2. This comment piece advocates training clinicians to use longitudinal biological age profiling to identify accelerated functional decline in younger adults2. Utilizing transparent, organ-specific aging models to recalibrate screening thresholds currently dominated by chronological age (CA)4,5. The challenge ahead is not to discard age entirely, but to contextualize it within a more precise and equitable model of care. The transition from chronological age to biological age (BA) estimations represents a paradigm shift in longevity medicine, moving away from what has been termed the “age illusion” toward a precision-based understanding of physiological reserve. While CA remains the standard for clinical triage and screening, it is an increasingly imprecise proxy for individual health status, often failing to account for the significant heterogeneity in aging trajectories among individuals of the same calendar age1. A critical advancement in BA estimation is the development of “causality-informed” epigenetic clocks. Unlike first-generation clocks that were purely correlative, newer models like DamAge distinguish between accumulated molecular damage and protective, adaptive responses. This distinction is vital for longevity medicine; a metric that only measures age-related changes without identifying their causal impact on mortality risk provides limited actionable data for therapeutic intervention3. By filtering for CpG sites that directly contribute to functional decline, these clocks offer a more robust framework for evaluating the efficacy of geroprotective treatments2,3.

However, the field is evolving beyond global BA toward organ-specific aging models. Recent research utilizing plasma protein-based signatures has demonstrated that organs age at distinct rates. These proteomic models have unveiled that accelerated aging of specific organismal systems—such as the cardiovascular or central nervous systems—can predict the onset of diseases like heart failure or dementia years before clinical symptoms manifest4. This approach allows for “healthy longevity medicine” to shift from reactive treatments to proactive, system-specific preservation.

The integration of these biomarkers into clinical practice is further supported by the rise of longevity biotechnology and artificial intelligence(AI)-driven platforms like Biolearn, which aim to standardize the interpretation of diverse biological datasets2. Yet, significant challenges remain. The medical community must address the “limitations of chronologic age” by recalibrating screening thresholds and clinical guidelines to reflect biological reality. This requires not only technological innovation but also a shift in medical education and ethics to ensure that BA-based decisions do not inadvertently lead to new forms of clinical inequity1. While BA estimations provide a sophisticated lens through which to view human aging, their clinical utility depends on the continued refinement of organ-specific proteomic signatures and the validation of causal epigenetic markers3,4. Moving forward, the synthesis of geroscience, artificial intelligence, and clinical applications will be essential to bridge the gap between experimental biomarkers and evidence-based longevity geriatric care6.

Artificial intelligence use for biological age estimations and health care

Notably, artificial intelligence is being developed and used for medical treatments, drug development, and robotic surgery, enhancing the effectiveness and safety of healthcare7. AI methods, including deep learning, graph neural networks, transformer-based architectures, and multimodal foundation models, are increasingly enabling the integration of longitudinal clinical data with multi-omic, imaging, wearable sensor, and environmental datasets to generate more accurate estimates of biological age and predict trajectories of healthy aging. By leveraging repeated measurements over time rather than single cross-sectional assessments, these approaches facilitate continuous monitoring of biological aging, early identification of accelerated aging, and dynamic evaluation of responses to lifestyle modifications and geroprotective interventions. In clinical practice, AI-driven biological age models have the potential to provide decision support by improving individualized risk stratification, guiding preventive and therapeutic interventions, and advancing precision longevity medicine through personalized management strategies that account for age-, sex-, and disease-specific characteristics. Moreover, AI and blockchain technologies have rapidly gained traction in many academic and professional domains8. AI-driven technologies in remote healthcare enhance patient care through improved diagnostics, monitoring, and medical treatment9. AI applications such as AI-enabled diagnostic systems, predictive analytics, and teleconsultation platforms overcome the limitations of traditional remote health care models9. The above-mentioned use of machine learning algorithms enables the analysis of vast amounts of patient data to predict health complications and tailor care plans10. AI is also changing how patients manage their medical conditions, particularly chronic diseases, through self-care tools, such as chatbots and health monitoring technologies11. AI-enabled virtual assistants can provide medication reminders and deliver health advice to older individuals, empowering patients and improving outcomes12,13,14,15,16. The integration of AI in healthcare promises to improve clinical decision-making, streamline operations, and enhance patient outcomes, which will be elucidated in the “Future Directions and Clinical Translation” below16. However, the deployment of AI in geriatric care and longevity medicine must adhere to robust ethical guidelines. In line with the Rome Call for AI Ethics17, developers and healthcare providers should ensure that AI systems are transparent, unbiased, inclusive, and secure, so that AI-driven interventions uphold human values and patient trust. By demanding explainability, fairness, and privacy protections in AI tools for aging populations, we can harness technology to augment care without compromising dignity or equity. The emergence of autonomous AI agents represents an important evolution beyond traditional AI systems in geriatric care. Unlike conventional AI tools that require continuous human oversight, AI agents can independently perform sequences of tasks, make contextual decisions, and adapt their behavior based on environmental feedback. These systems can coordinate care across multiple providers, manage complex medication schedules, and provide adaptive conversational support that responds to individual needs. However, their autonomous nature raises distinct ethical concerns regarding accountability, the risk of reducing human contact, and the need for transparency in decision-making. Ethically responsible implementation of AI-driven biological age assessment requires robust governance to mitigate algorithmic bias, ensure model transparency and explainability, safeguard individual ownership and privacy of biological age data, and prevent its misuse in employment or insurance underwriting, where inaccurate or inequitable risk stratification could exacerbate existing health disparities. Furthermore, meaningful informed consent should extend beyond routine data collection to include clear communication regarding the intended use, limitations, potential downstream consequences, and data-sharing practices associated with biological age prediction, thereby preserving patient autonomy and fostering public trust in precision longevity medicine.

Limitations, future directions, and clinical translation

Despite remarkable advances in the development of molecular biomarkers of aging, the accurate quantification of biological age remains an unresolved scientific and clinical challenge. Although epigenetic clocks currently represent the most extensively validated molecular tools, biological aging is a multidimensional process that extends beyond epigenetic remodeling18. Mitochondrial dysfunction, chronic low-grade inflammation, alterations of the gut microbiome, cellular senescence, metabolic dysregulation, and immune remodeling all contribute to aging trajectories and should be integrated into future multimodal aging algorithms19. Consequently, reliance on a single molecular layer is unlikely to capture the full biological complexity underlying individual aging processes.

An additional limitation is the absence of robust organ-specific biomarkers of biological age. Direct assessment of tissue-specific aging frequently requires invasive sampling, limiting its feasibility for routine clinical practice. As a result, many current studies infer accelerated organ aging from the presence of chronic organ-specific diseases. While clinically pragmatic, this approach risks circular reasoning by equating disease burden with biological aging rather than independently measuring aging itself. Future efforts should therefore prioritize the development of minimally invasive biomarkers capable of accurately quantifying organ-specific biological age before overt disease develops.

Artificial intelligence and machine learning have substantial potential to transform precision longevity medicine, but their performance fundamentally depends on the quality and biological relevance of the input data20. Similar to advances achieved in structural biology, successful AI applications require carefully designed biological representations rather than simply increasing data volume. Aging is inherently a dynamic spatiotemporal process involving complex interactions across molecular, cellular, tissue, and organismal levels. Therefore, longitudinal multimodal datasets incorporating temporal trajectories, environmental exposures, imaging, clinical phenotypes, multi-omics, and functional measurements will be essential for developing biologically meaningful predictive models. In this context, deep phenotyping represents a critical prerequisite for future AI-driven aging prediction, enabling comprehensive characterization of individual aging trajectories rather than static snapshots of chronological age.

Precision longevity medicine must explicitly incorporate sex-, gender-, and ethnicity-specific determinants of aging, as biological aging trajectories are profoundly influenced by hormonal transitions such as menopause, and genetic ancestry21. Importantly, sociocultural factors collectively modify the risk of dementia, cardiovascular disease, and other age-related conditions and should not be neglected either. Integrating these variables into multimodal biological age models will improve risk stratification, enable more personalized preventive interventions, and promote equitable implementation of precision medicine across diverse populations21. Besides, biological aging differs substantially between women and men owing to differences in endocrine regulation, immune responses, mitochondrial function, reproductive aging, and susceptibility to age-related diseases. Gender-related behavioral, socioeconomic, and environmental factors further modify aging trajectories and health outcomes. Future precision longevity models should therefore explicitly incorporate both biological sex and gender-related variables to improve predictive accuracy, therapeutic stratification, and equitable implementation across diverse populations. From a clinical perspective, translation of biological age assessment into routine medical practice will require standardized measurement platforms, external validation across ethnically and geographically diverse populations, harmonized regulatory frameworks, and prospective interventional studies demonstrating that modification of biological age translates into improved clinical outcomes. Rather than serving solely as prognostic biomarkers, biological age metrics should ultimately support individualized prevention strategies, guide selection of geroprotective interventions, optimize screening intervals, and enable longitudinal monitoring of therapeutic responses within precision longevity medicine.

Conclusions

In summary, the above-mentioned developments underscore that future biological age assessment will likely evolve toward an integrated, systems-based framework that combines molecular biomarkers, advanced imaging, functional testing, digital health technologies, and artificial intelligence. In consequence, it will be possible to generate clinically actionable measures of aging that can inform personalized healthcare throughout the life course and be individualized for women and men separately, respecting their ethnicity. This article aimed to shed light on how AI accelerates the clinical implementation of biological age rather than chronological age in precision medicine and to stimulate new research on this exciting topic.

Data availability

No datasets were generated or analysed during the current study.

References

  1. Lee, M. G. The age illusion - limitations of chronologic age in medicine. N. Engl. J. Med. 394, 1251–1253 (2026).

    Article  PubMed  Google Scholar 

  2. Lyu, Y. X. et al. Longevity biotechnology: bridging AI, biomarkers, geroscience and clinical applications for healthy longevity. Aging 16, 12955–12976 (2024).

    PubMed  PubMed Central  Google Scholar 

  3. Ying, K. Causal inference for epigenetic ageing. Nat. Rev. Genet. 26, 3 (2025).

    Article  PubMed  Google Scholar 

  4. Goeminne, L. J. E. et al. Plasma protein-based organ-specific aging and mortality models unveil diseases as accelerated aging of organismal systems. Cell Metab. 37, 205–222.e6 (2025).

    Article  PubMed  Google Scholar 

  5. Zhang, B. et al. Epigenetic clocks and longitudinal plasma biomarkers of Alzheimer’s disease. Alzheimers Dement 21, e70983 (2025).

    Article  PubMed  PubMed Central  Google Scholar 

  6. Jones, C. H. & Dolsten, M. Healthcare on the brink: navigating the challenges of an aging society in the United States. npj Aging 10, 22 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

  7. Furlepa, K. et al. Recommendations for the development of telemedicine in Poland based on the analysis of barriers and selected telemedicine solutions. Int. J. Environ. Res. Public Health 19, 1221 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

  8. Thacharodi, A. et al. Revolutionizing healthcare and medicine: the impact of modern technologies for a healthier future—a comprehensive review. Health Care Sci. 3, 329–349 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

  9. Chaturvedi, U., Chauhan, S. B. & Singh, I. The impact of artificial intelligence on remote healthcare: enhancing patient engagement, connectivity, and overcoming challenges. Intell. Pharm. 3, 323–329 (2025).

  10. Li, Y.-H., Li, Y.-L., Wei, M.-Y. & Li, G.-Y. Innovation and challenges of artificial intelligence technology in personalized healthcare. Sci. Rep. 14, 18994 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

  11. Mennella, C., Maniscalco, U., De Pietro, G. & Esposito, M. Ethical and regulatory challenges of AI technologies in healthcare: a narrative review. Heliyon 10, e26297 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

  12. Dave, M. & Patel, N. Artificial intelligence in healthcare and education. Br. Dent. J. 234, 761–764 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  13. Briganti, G. & Le Moine, O. Artificial intelligence in medicine: today and tomorrow. Front. Med. 7, 27 (2020).

    Article  Google Scholar 

  14. Al Kuwaiti, A. et al. A review of the role of artificial intelligence in healthcare. J. Pers. Med. 13, 951 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  15. Ferrara, M. et al. Risk management and patient safety in the artificial intelligence era: a systematic review. Healthcare 12, 549 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

  16. Charow, R. et al. Artificial intelligence education programs for health care professionals: scoping review. JMIR Med. Educ. 7, e31043 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  17. Pontifical Academy for Life. Rome Call for AI Ethics. Vatican City. https://www.vatican.va/roman_curia/pontifical_academies/acdlife/documents/rc_pont-acd_life_doc_20202228_rome-call-for-ai-ethics_en.pdf (2020).

  18. Wyss-Coray, T. & Topol, E. J. Biological aging clocks in health and disease. Nat. Med. https://doi.org/10.1038/s41591-026-04495-3 (2026).

  19. Ji, Q. et al. Multimodal omics approaches to aging and age-related diseases. Phenomics 4, 56–71 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

  20. Green, J. B. & Haykal, D. AI and longevity medicine: unlocking predictive and preventive strategies for healthy aging. Presse Med. 55, 10435 (2026).

    Article  Google Scholar 

  21. Hägg, S. & Jylhävä, J. Sex differences in biological aging with a focus on human studies. eLife 10, e63425 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The author used Grammarly to enhance the accuracy of their English and improve the flow of their writing.

Funding

P.K. received the OPUS 28 grant, funded by the National Science Center, grant number OPUS28/2024/55/B/NZ9/02063, and IDUB POB grant, funded by the Nicolaus Copernicus University, grant number FemLife_Omics.

Author information

Authors and Affiliations

  1. Department of Basic and Preclinical Sciences, Nicolaus Copernicus University, Torun, Poland

    Pawel Kordowitzki

  2. Institute of Advanced Studies, Nicolaus Copernicus University, Torun, Poland

    Pawel Kordowitzki

  3. Department of Gynecology with Center for Oncological Surgery, Campus Virchow Klinikum, Charité-Universitätsmedizin Berlin, Berlin, Germany

    Pawel Kordowitzki

Authors

  1. Pawel Kordowitzki

Contributions

P.K.: manuscript writing, editing, conceptualization, and critical discussion.

Corresponding author

Correspondence to Pawel Kordowitzki.

Ethics declarations

Competing interests

The author declares no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kordowitzki, P. The use of biological age and artificial intelligence in health care. npj Aging 12, 147 (2026). https://doi.org/10.1038/s41514-026-00524-0

Download citation

  • Received: 04 May 2026

  • Accepted: 08 September 2026

  • Published: 26 October 2026

  • Version of record: 26 October 2026

  • DOI: https://doi.org/10.1038/s41514-026-00524-0

来源:Nature 健康护理主题 · nature.com