AI-powered video tool could detect high blood pressure, diabetes

Credit: Unsplash+.

A new AI-driven tool that uses high-speed video to detect subtle changes in blood flow across the face and hands may offer a revolutionary way to screen for high blood pressure and diabetes without needing blood pressure cuffs, blood tests, or wearable devices.

Developed by researchers in Japan, this contact-free system was tested on patients in a hospital setting and demonstrated high accuracy, potentially allowing people to monitor these health conditions from home.

Presented at the American Heart Association’s Scientific Sessions, the findings suggest this tool could detect early signs of hypertension and diabetes with minimal effort.

According to the study’s lead researcher, Ryoko Uchida from the University of Tokyo, the technology could enable individuals to monitor their blood pressure and blood sugar levels in a “noninvasive, non-contact and non-active manner” without having to wear devices or visit a hospital.

The system captures blood flow changes using a high-speed video camera set to 150 frames per second. This footage is then analyzed by a machine learning algorithm, which examines pulse wave arrival times at specific points on the face and hands.

In the study, 215 participants, including those with known hypertension and diabetes and healthy individuals for comparison, sat still while video recordings were taken.

The algorithm measured pulse waves in 30 regions of interest, 22 on the face and eight on the hands, to assess blood flow characteristics.

The results were promising: in detecting stage 1 hypertension (systolic blood pressure of 130-139 mmHg or diastolic pressure of 80-89 mmHg), the system was 94% accurate in a subgroup of 77 participants.

For diabetes, measured against the hemoglobin A1C test (a standard test for long-term blood glucose levels), the system achieved a 75% accuracy rate in identifying those with an A1C of 6.5 or higher, the threshold for diabetes.

Despite these findings, researchers acknowledge that the system is still in its early development stages.

Uchida noted that additional studies are needed to determine if the technology performs accurately in different environments, such as brighter or darker settings, and with populations outside Japan.

However, not all experts are convinced of the tool’s practicality.

Dr. Geoff Rubin, a medical imaging expert at the University of Arizona, pointed out that while contactless testing is innovative, the system still requires people to sit in a controlled environment, which may not be significantly easier than traditional methods.

“Why not just use a blood pressure cuff or take a small blood sample in the same amount of time?” Rubin asked, emphasizing that although the tool could one day prove useful, there are still many unanswered questions.

While the tool shows potential to enhance screening accessibility, future studies will need to address its limitations and ensure its effectiveness in a wider variety of settings.

Nonetheless, this technology represents a step toward simpler, more accessible health monitoring, especially for those who may not engage with wearables or traditional testing methods.

If you care about blood pressure, please read studies about unhealthy habits that could increase high blood pressure risk, and eating eggs in a healthy diet may reduce risks of diabetes, high blood pressure.

For more information about blood pressure, please see recent studies that early time-restricted eating could help improve blood pressure, and results showing 12 foods that lower blood pressure.

Copyright © 2024 Knowridge Science Report. All rights reserved.