SINGAPORE: Artificial intelligence (AI) is poised to revolutionise ageing research and improve the quality of life for older adults, according to a collaborative study conducted by researchers from the Yong Loo Lin School of Medicine at the National University of Singapore (NUS Medicine) and the Institute for Biostatistics and Informatics in Medicine and Ageing Research at Rostock University Medical Center in Germany.
The study, published in Ageing Research Reviews, explores how advanced AI tools, including Large Language Models (LLMs), can streamline the evaluation of ageing-related interventions and offer personalised health recommendations.
As research into ageing continues to generate vast amounts of data, scientists face challenges in determining which interventions—such as new pharmaceuticals, dietary modifications, or exercise regimens—are safe and effective. The study examined how AI can be harnessed to analyse this information more efficiently and accurately. It also proposed a set of standards to ensure AI-generated evaluations are reliable, precise, and easily interpretable.
The researchers outlined eight key criteria that AI systems must meet to provide effective evaluations. These include ensuring the correctness of results, maintaining high data quality, and offering useful and comprehensive assessments. AI-generated insights should be clear, concise, and explainable while taking into account causal mechanisms influenced by interventions.
The study also highlighted the importance of evaluating data holistically, considering both efficacy and toxicity and ensuring analyses are conducted in interdisciplinary settings. Reproducibility, standardisation, and harmonisation of AI analyses were emphasised, along with a focus on large-scale, diverse longitudinal data and known ageing mechanisms.
By incorporating these requirements into AI prompting, the researchers observed significant improvements in the quality of AI-generated recommendations. Professor Brian Kennedy, from the Department of Biochemistry & Physiology and the Healthy Longevity Translational Research Programme at NUS Medicine, who co-led the study, explained how AI was tested using real-world examples, including medicines and dietary supplements.
He noted that when assessing rapamycin, a drug often studied for its potential role in healthy ageing, AI evaluated its effectiveness and provided nuanced explanations, including potential side effects and limitations.
Professor Georg Fuellen, Director of the Institute for Biostatistics and Informatics in Medicine and Ageing Research at Rostock University Medical Center and co-lead of the study, highlighted the broader implications of their findings. He noted that ensuring AI adheres to critical evaluation standards could enhance its ability to identify safer and more effective treatments.
Additionally, AI-driven insights could improve clinical trial design and enable more personalised health recommendations, benefiting individuals as they age.
The study marks a significant step towards leveraging AI to optimise healthcare strategies and improve ageing outcomes, demonstrating the potential for AI to support healthier, longer lives.