Deep aging clocks: AI-powered strategies for biological age estimation

IF 12.4 1区医学 Q1 CELL BIOLOGY

Ageing Research Reviews Pub Date : 2025-09-02 DOI:10.1016/j.arr.2025.102889

Luma Srour , Yosra Bejaoui , James She , Tanvir Alam , Nady El Hajj

{"title":"Deep aging clocks: AI-powered strategies for biological age estimation","authors":"Luma Srour , Yosra Bejaoui , James She , Tanvir Alam , Nady El Hajj","doi":"10.1016/j.arr.2025.102889","DOIUrl":null,"url":null,"abstract":"<div><div>Several strategies have emerged lately in response to the rapid increase in the aging population to enhance health and life span and manage aging challenges. Developing such strategies is imperative and requires an assessment of biological aging. Several aging clocks have recently been developed to measure biological aging and to assess the efficacy of longevity interventions. Biological age better reflects a person’s actual age and is closely associated with health outcomes and time to mortality. Traditionally, most aging clocks assume that biological changes occur linearly over time. However, age-related changes do not necessarily follow a linear trajectory. Thus, “Deep Aging Clocks” have been developed to overcome previous clocks' limitations and better capture subtle changes that occur during aging. Here, we summarize the current deep aging clocks, including epigenetics, transcriptomics, metabolomics, microbiome, and imaging based clocks for age prediction. Recent advances in artificial intelligence (AI), utilizing deep learning techniques, have significantly enhanced the prediction of biological aging, and this would help improve aging clocks and accelerate efforts to reach longer and healthier lives.</div></div>","PeriodicalId":55545,"journal":{"name":"Ageing Research Reviews","volume":"112 ","pages":"Article 102889"},"PeriodicalIF":12.4000,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ageing Research Reviews","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1568163725002351","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CELL BIOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Several strategies have emerged lately in response to the rapid increase in the aging population to enhance health and life span and manage aging challenges. Developing such strategies is imperative and requires an assessment of biological aging. Several aging clocks have recently been developed to measure biological aging and to assess the efficacy of longevity interventions. Biological age better reflects a person’s actual age and is closely associated with health outcomes and time to mortality. Traditionally, most aging clocks assume that biological changes occur linearly over time. However, age-related changes do not necessarily follow a linear trajectory. Thus, “Deep Aging Clocks” have been developed to overcome previous clocks' limitations and better capture subtle changes that occur during aging. Here, we summarize the current deep aging clocks, including epigenetics, transcriptomics, metabolomics, microbiome, and imaging based clocks for age prediction. Recent advances in artificial intelligence (AI), utilizing deep learning techniques, have significantly enhanced the prediction of biological aging, and this would help improve aging clocks and accelerate efforts to reach longer and healthier lives.

查看原文本刊更多论文

深度老化时钟：人工智能驱动的生物年龄估计策略

最近出现了一些战略，以应对老龄化人口的迅速增加，以提高健康和寿命，并应对老龄化挑战。制定这样的策略势在必行，需要对生物老化进行评估。最近已经开发了几种衰老时钟来测量生物衰老和评估长寿干预措施的功效。生物年龄更好地反映了一个人的实际年龄，并与健康状况和死亡时间密切相关。传统上，大多数衰老时钟假设生物变化随时间线性发生。然而，与年龄相关的变化并不一定遵循线性轨迹。因此，“深度老化时钟”已经开发出来，以克服以前时钟的局限性，更好地捕捉衰老过程中发生的细微变化。在这里，我们总结了目前的深度衰老时钟，包括表观遗传学、转录组学、代谢组学、微生物组学和基于成像的年龄预测时钟。利用深度学习技术的人工智能（AI）的最新进展大大增强了对生物衰老的预测，这将有助于改善衰老时钟，加速实现更长寿、更健康的生活。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Ageing Research Reviews 医学-老年医学

CiteScore

19.80

自引率

2.30%

发文量

216

审稿时长

55 days

期刊介绍： With the rise in average human life expectancy, the impact of ageing and age-related diseases on our society has become increasingly significant. Ageing research is now a focal point for numerous laboratories, encompassing leaders in genetics, molecular and cellular biology, biochemistry, and behavior. Ageing Research Reviews (ARR) serves as a cornerstone in this field, addressing emerging trends. ARR aims to fill a substantial gap by providing critical reviews and viewpoints on evolving discoveries concerning the mechanisms of ageing and age-related diseases. The rapid progress in understanding the mechanisms controlling cellular proliferation, differentiation, and survival is unveiling new insights into the regulation of ageing. From telomerase to stem cells, and from energy to oxyradical metabolism, we are witnessing an exciting era in the multidisciplinary field of ageing research. The journal explores the cellular and molecular foundations of interventions that extend lifespan, such as caloric restriction. It identifies the underpinnings of manipulations that extend lifespan, shedding light on novel approaches for preventing age-related diseases. ARR publishes articles on focused topics selected from the expansive field of ageing research, with a particular emphasis on the cellular and molecular mechanisms of the aging process. This includes age-related diseases like cancer, cardiovascular disease, diabetes, and neurodegenerative disorders. The journal also covers applications of basic ageing research to lifespan extension and disease prevention, offering a comprehensive platform for advancing our understanding of this critical field.