Researchers at Johns Hopkins University have discovered a more accurate method of identifying patients’ obesity risk using artificial intelligence and machine learning, providing an alternative to the long-used body mass index (BMI),
Unlike BMI, which majorly relies on calculations using a person’s weight and height, the new AI tool can predict a person’s waistline by analyzing their age, weight, height, ethnicity, and level of education with remarkable accuracy, ushering in a new era of obesity risk determination by doctors.
With this tool, clinicians can now predict patients’ risk of obesity-related diseases such as heart disease, diabetes, stroke, and other diseases usually assessed using BMI.
Determining the Risk of Obesity Based on Waistline
Although waist circumference is often associated with the risk of obesity and health conditions such as diabetes, and cardiovascular illnesses, it’s not regularly measured when one visits clinics.
However, with this advanced waist measurement tool, developed by engineers at Johns Hopkins’ Artificial Intelligence for Engineering and Medicine Lab, one does not need physical waist measurement as the AI-powered tool can make accurate predictions based on other parameters such as age, weight, ethnicity, and level of education.
This can help doctors quickly asses patients’ risk of suffering from health conditions related to obesity, saving time and improving the accuracy of assessment.
The study, which was led by Carl Harris, a biomedical engineering student, and Prasanna Santhanam, an associate professor in the Division of Endocrinology, Diabetes, and Metabolism, with contributions from a biomedical engineering student Daniel Olshvang, underscores the potential of artificial intelligence in the treatment of obesity.
Additionally, the study demonstrates using waist circumference for predicting risks of obesity-related health conditions is better than BMI. However, the lack of a standardized measuring technique and its rare use in clinics are some of the challenges with this new approach.
Image credit: Image by ennrick from Pixabay
