Preventing Aging and Chronic Disease Risks Measured by Artificial Intelligence

Professor Kwon Oh-ran's Team at Ewha Womans University Develops and Validates Oxidative Stress Risk Assessment Model and Personalized Nutrition Provision Technology

Reference photo. Not related to the article.

[Asia Economy Reporter Kim Bong-su] A technology has been developed that uses artificial intelligence and big data to measure a patient's oxidative stress risk and provide appropriate nutrition to prevent aging and various chronic diseases. Oxidative stress refers to a state in which harmful oxygen species are excessively generated in the body due to excessive stress, ultraviolet rays, smoking and heavy drinking, automobile exhaust, and excessive consumption of instant foods, and it causes aging or chronic diseases.

The National Research Foundation of Korea announced on the 12th that an international joint research team led by Professor Kwon Oh-ran of Ewha Womans University developed and validated a machine learning model to diagnose oxidative stress risk in Koreans.

The advancement of big data analysis and artificial intelligence is helping to provide suitable medical information to patients or reduce diagnostic and treatment errors based on large amounts of cohort data (data from the same group) and clinical information. However, there are very few studies that apply machine learning algorithms to reduce the risk of oxidative stress-related chronic diseases in advance and provide precision nutrition.

The research team aimed to solve this problem through a proof-of-concept study to test the application of spatial methodology to extensive human subject research. They used machine learning methods to develop a model that determines oxidative risk with strong predictive power and interpretability by considering the characteristics of variables in a spectrum of diverse and complex data. They developed a predictive model that comprehensively considers 16 variables such as age, BMI, diet quality, and blood markers to quantify the user's oxidative stress risk level.

The research team validated this model using data from 2,454 individuals who visited Boramae Hospital in Seoul for health check-ups between April 2015 and August 2018. The sensitivity was 0.923 (95% CI: 0.879-0.967), and the accuracy was 0.891 (95% CI: 0.854-0.928).

The research team plans to continue studies related to developing quantitative models for chronic diseases by validating the developed model using national cohort data, comparing and analyzing data from various ethnic and national populations.

This research is expected to help establish health management strategies that can prevent chronic diseases related to diet and lifestyle by presenting a model that stratifies and predicts oxidative stress risk in healthy populations.

The research results were published on July 16 in the international antioxidant journal Antioxidants.

This content was produced with the assistance of AI translation services.