summary: A new study used machine learning to identify lifestyle and health factors that are most strongly associated with cognitive performance across life expectancy. Of the 374 adults aged 19-82 years, age, blood pressure, and BMI were top predictors of success in focus and speed-based attention tests.
Diet and exercise played a smaller role, but they were still associated with better outcomes, especially when offsetting high BMI or blood pressure. This data-driven approach emphasizes combining multiple factors to make brain health more clear about age-related factors.
Important facts:
- Top predictor: Age, diastolic blood pressure, and BMI most strongly influenced cognitive performance.
- Diet + Exercise: Healthy diet and physical activity contributed modestly but positively to the focus and speed of response.
- Benefits of machine learning: Advanced algorithms revealed nuanced relationships that traditional statistics may overlook.
sauce: University of Illinois
The new research will provide insights into health and lifestyle indicators, including diet, physical activity and weight, which are most closely aligned with healthy brain function throughout life.
In this study, machine learning was used to determine which variables best predicted their ability to quickly complete tasks without distraction.
It has been reported in Nutritional Journalthis study found that age, blood pressure, and body mass index were the most powerful predictors of success in a test called the flanker task.
Diet and exercise also played a smaller but relevant role in test performance.
“This study used machine learning to evaluate many variables at once to help identify the variables that are most closely aligned with cognitive performance,” said Nyman Khan, professor of health and kinesiology at the University of Illinois Urbana-Champaign, who led the work of Dr. Kinesiology. Student Shreya Berma.
“Standard statistical approaches cannot accept this level of complexity at once.”
To build the model, the team used data collected from 374 adults aged 19 to 82. Data included participants’ demographics such as age, BMI, blood pressure, and physical activity level, as well as dietary patterns and performance from the flanker test, which measured processing speed and accuracy when determining the orientation of the central arrows where other arrows pointed in the same or opposite directions were pinched.
“This is an established measure of cognitive function that assesses attention and inhibitory control,” Khan said.
Previous studies have found that several factors are related to cognitive conservation throughout life span, Khan said.
“Adhering to the Healthy Diet Index, a measure of diet quality, is linked to superior executive functioning and processing speed in older people,” he said. “Other studies have found that diets rich in antioxidants, omega-3 fatty acids and vitamins are associated with improving cognitive function.”
The dietary approaches to stop eating a combination of two known as the high blood pressure, or dash, Mediterranean diet, and mind diet, are all “related to protective effects against cognitive decline and dementia,” the researchers wrote. Physical factors such as BMI and blood pressure and increased physical activity are also powerful predictors of cognitive health or reduction in aging.
“Obviously, cognitive health is driven by a number of factors, but which one is most important?” Verma said. “We wanted to assess the relative strength of each of these factors in combination with everything else.”
Machine learning “provides a promising tool for analyzing large datasets with multiple variables and identifying patterns that may not be revealed through traditional statistical approaches,” the researchers write.
The team tested a variety of machine learning algorithms to optimally examine a variety of factors to predict the accurate response speed of flanker tests. Researchers tested the predictive capabilities of each algorithm and used a variety of approaches to verify what appeared to perform best.
They found that age was the most influential predictor of performance on the test, with persistent diastolic blood pressure, BMI and systolic blood pressure. Compliance with a healthy feeding index predicted less cognitive performance than blood pressure or BMI, but was also correlated with improved test performance.
“Physical activity emerges as a moderate predictor of reaction time, and the results suggest that it may interact with other lifestyle factors, such as diet and weight, to affect cognitive performance,” Khan said.
“This research reveals how machine learning can bring accuracy and nuance to the field of nutritional neuroscience,” he said.
“By moving beyond traditional approaches, machine learning can help coordinate strategies for aging populations, individuals at metabolic risk, or those seeking to enhance cognitive function through lifestyle changes.”
I. of I. Personalized Nutrition Initiative and the National Center for Supercomputing Applications supported this study.
Kahn is a nutritionist and affiliate professor at Illinois Department of Nutrition Science, Neuroscience Programs and Beckman Institute of Advanced Science and Technology.
About this AI and Brain Health Research News
author: Diana Yates
sauce: University of Illinois
contact: Diana Yates – University of Illinois
image: This image is credited to Neuroscience News
Original research: Open access.
“Predict cognitive outcomes through nutrition and health markers using monitored machine learning“Written by Nyman Khan et al. Journal of Nutrition
Abstract
Predict cognitive outcomes through nutrition and health markers using monitored machine learning
background
Although the use of machine learning (ML) in health research is growing, applications for predicting cognitive outcomes using a variety of health indicators are underinvestment.
the purpose
We aim to use the ML model to predict cognitive performance based on a set of health and behavioral factors and to identify key contributors to cognitive functioning for insights into potential personalized interventions.
method
Using data from 374 adults aged 19-82 (227 women), we developed an ML model that predicts cognitive performance (reaction time in milliseconds) of modified Eriksen Franker tasks.
Characteristics include demographics, anthropometric measurements, dietary index (a healthy diet index, dietary approach to stop hypertension, a discipline approach from the Mediterranean and the Mediterranean to stop hypertension interventions for neurodegeneration delay), self-reported physical activity, and systolic and diastolic blood pressure. The dataset was split for training and testing (80:20).
Predictive models (decision tree, random forest, adabost, xgboost, slope boost, linear, ridge, and lasso regression) were used for hyperparameter adjustment and cross-validation. The importance of features was calculated using permutation importance, while performance was calculated using mean absolute error (MAE) and mean square error.
result
The Random Forest Regress showed the best performance with the lowest MAE (training: 0.66ms, test: 0.78ms) and average square error (training: 0.70ms).2;Test: 1.05ms2). Age was the most important trait (score: 0.208), followed by diastolic blood pressure (0.169), BMI (0.079), systolic blood pressure (0.069), and healthy feeding index (0.048). Ethnicity (0.005) and gender (0.003) had minimal predictive effect.
Conclusion
Age, blood pressure, and BMI are strongly associated with cognitive performance, but dietary quality has subtle effects. These findings highlight the potential of ML models to develop individualized interventions and prevention strategies for cognitive decline.
