A cross-disciplinary team from the University of Alberta is utilizing machine learning to predict the future mental and physical health of aging Canadians. Led by Bo Cao, an associate professor of psychiatry and Canada Research Chair in Computational Psychiatry, the team is developing machine learning programs that analyze health-related, lifestyle, socio-economic, and other data to provide individualized care and promote healthy aging.
The team’s approach involves using rich de-identified data from the Canadian Longitudinal Study on Aging (CLSA), which follows more than 30,000 Canadians aged 45 to 85 over a span of up to 25 years. In a recently published study in the journal Gerontology, the team developed a biological age index by applying machine learning models to blood test data from the CLSA. This index, known as the BioAge gap, compares physiological age to chronological age and provides insights into individual aging processes.
The researchers discovered strong associations between a positive BioAge gap (indicating an older physiological age) and chronic illness, frequent consumption of processed and red meat, smoking, and passive smoke exposure. On the other hand, negative BioAge gaps (indicating a younger physiological age) were associated with factors such as consumption of fruits, legumes, and vegetables.
In a separate study published in the Journal of Affective Disorders, the team developed a machine learning program that accurately predicted depression onset within three years. The model, trained using records of individuals eventually diagnosed with depression, identified factors such as subthreshold depression symptoms, emotional instability, low life satisfaction, perceived health and social support, and nutrition risk as key predictors of depression.
Although the machine learning models are not yet refined enough for real-world implementation, Cao and his team hope to refine them further in order to provide personalized care and contribute to the health and well-being of aging individuals. The goal is to create a healthy aging trajectory for all. By understanding the associations and risk factors for aging, effective public health recommendations can be made to promote healthy longevity and prevent the onset of diseases such as depression.
Through their research, the University of Alberta team aims to establish a dialogue that involves clinicians, patients, and individuals with lived experience to ensure these machine learning models can benefit the general public. While more research and testing are planned, Cao and his team are optimistic about the potential of machine learning to revolutionize healthcare and support healthy aging.
Overall, machine learning holds great promise in predicting the future mental and physical health of aging Canadians. By leveraging data-driven insights, healthcare teams can provide individualized care and interventions, ultimately promoting healthy aging and improving the well-being of older adults.
