Researchers at the Institute of Psychiatry, Psychology & Neuroscience (IoPPN) at King’s College London have undertaken a significant study examining the use of artificial intelligence to create aging clocks that predict health and lifespan through blood data analysis. The research, seen as a pioneering effort in its field, reflects the growing intersection of artificial intelligence and healthcare.
The study evaluated 17 different machine learning algorithms using extensive data from 225,000 participants of the UK Biobank, all aged between 40 and 69 years at the time of recruitment. The focus was on how effectively various metabolomic aging clocks can forecast lifespan and their reliability in association with health and aging metrics.
A key concept introduced in the study is a person's "MileAge," which signifies their metabolomic age—the internal age of their body based on blood markers known as metabolites. These metabolites are by-products of metabolic processes, such as the breakdown of food into energy. The study calculated the "MileAge delta," a difference between the metabolite-predicted age and chronological age, to determine whether an individual’s biological ageing is accelerating or slowing down.
Published in the journal Science Advances, this research is notable for being the first comprehensive comparison of different machine learning approaches to developing biological aging clocks using metabolomic data, leveraging one of the largest datasets globally. The project was funded by the National Institute for Health and Care Research (NIHR) and utilized resources from the UK Biobank.
Key findings indicated that individuals exhibiting accelerated biological ageing—those with a metabolite-predicted age exceeding their chronological age—were generally frailer. These individuals reported poorer health, exhibited a greater likelihood of chronic illnesses, and faced higher mortality risks. Such participants also had shorter telomeres, which are associated with age-related diseases like atherosclerosis. Conversely, those with decelerated biological ageing—having a metabolite-predicted age younger than their chronological age—were only weakly linked to improved health outcomes.
The researchers propose that these aging clocks may serve as tools for identifying early signs of declining health, potentially enabling preventive strategies and interventions before the onset of diseases. They may also empower individuals to monitor their health, make informed lifestyle choices, and take proactive measures to enhance their wellbeing over time.
Dr. Julian Mutz, a King’s Prize Research Fellow at the IoPPN and the study's lead author, stated: “Metabolomic ageing clocks have the potential to provide insights into who might be at greater risk of developing health problems later in life. Unlike chronological age, which cannot be changed, our biological age is potentially modifiable.” He added that the aging clocks could aid in shaping lifestyle choices and informing preventative strategies in health services.
Professor Cathryn Lewis, a Professor of Genetic Epidemiology & Statistics and Co-Deputy Lead of the Trials, Genomics and Prediction theme at the NIHR Maudsley BRC, expressed that there is considerable interest in accurately developing these aging clocks, noting that “powerful big data analytics can play a critical role in advancing these tools.” She described the study as a significant step towards establishing the viability of biological aging clocks in influencing health decisions.
The research identified that among the various algorithms, one developed through a method known as Cubist rule-based regression demonstrated the strongest correlation with health and ageing markers. Furthermore, algorithms capable of modelling non-linear relationships between metabolites and age were found to be particularly effective in capturing signals related to health and longevity.
As developments in artificial intelligence continue to shape the landscape of healthcare, studies like this underscore the transformative potential of these technologies in understanding and managing biological ageing processes.
Source: Noah Wire Services