Dementia encompasses a range of debilitating neurological conditions marked by progressive memory loss and cognitive decline. Currently, over 55 million people globally are affected by dementia, with projections indicating this number will rise in the coming decades.
Due to limited effective treatments for dementia, researchers have focused on identifying factors that contribute to its onset, aiming to develop preventive measures or interventions that reduce dementia risk.
A recent collaboration between Fudan University, Zhejiang University School of Medicine, and other institutions employed machine learning to design a dietary intervention intended to lower the risk of late-life dementia. This new approach, named MODERN (Machine-learning-assisted Optimizing Dietary intERvention against demeNtia risk), was presented in Nature Human Behavior.
Nature Human Behaviour"The scarcity of effective dementia prevention strategies prompted our research," explained Prof. Jintai Yu, the paper's corresponding author, speaking to Medical Xpress. "Although dietary modifications offer a promising risk factor for intervention, existing approaches often fall short in demonstrating significant benefits through randomized controlled trials (RCTs). Our goal was to create a novel data-driven dietary strategy that leverages large-scale cohort data and advanced machine learning techniques to reduce dementia risk."
Utilizing machine learning, Prof. Yu and his team analyzed data from the UK Biobank to identify optimal food combinations linked with decreased dementia risk. The researchers employed LightGBM, a machine learning algorithm trained on dietary and health data from 185,012 individuals in the UK.
"Out of the algorithms we tested—including XGBoost and Random Forest—LightGBM performed best, achieving the highest area under the ROC curve (AUC)," noted Prof. Jia You, co-first author of the study. "This model identified key dietary components associated with dementia risk. These insights were transformed into a practical scoring system called MODERN, which emphasizes consuming moderate amounts of brain-beneficial foods like leafy greens and berries while restricting harmful items such as sweetened beverages."
The newly developed MODERN diet was found to correlate more strongly with reduced dementia risk than other established dietary programs focused on brain health. Specifically, it showed greater protective effects compared to the MIND (Mediterranean Intervention for Neurodegenerative Delay) diet.
"Participants scoring highest in the MODERN diet exhibited a 36% lower dementia risk across three independent external validation cohorts," stated Dr. Sijia Chen, the study’s first author. "Mechanistic analyses uncovered potential neuroprotective pathways, such as enhanced brain structural integrity and reduced neuroinflammation."
Future assessments of the MODERN diet's protective effects in real-world settings may validate its benefits further. Ultimately, it could be incorporated into public health initiatives to help reduce neurodegenerative diseases worldwide.
"Moving forward, we plan to validate the MODERN diet across diverse populations and conduct randomized controlled trials (RCTs) to confirm its causal preventive benefits," added Prof. Yu. "Furthermore, we intend to apply similar data-driven methods to identify optimal dietary patterns for other brain-related disorders such as anxiety and depression. Our long-term objective is to develop a comprehensive, evidence-based dietary framework designed to promote brain health and prevent neurological diseases."
This article was crafted by our author Ingrid Fadelli, edited by Gaby Clark, and fact-checked and reviewed by Andrew Zinin—demonstrating the power of careful human effort. We depend on readers like you to keep independent science journalism vibrant. If this content resonates with you, please consider a donation, especially a monthly one. In recognition, you'll receive an ad-free experience.