AHA 2024: AI Tool Could Provide Fast, No-Contact Blood Pressure and Diabetes Screenings

Results from a preliminary study presented at the American Heart Association (AHA) 2024 Scientific Sessions suggests that an artificial intelligence (AI)-powered algorithm using a high-speed, short video of skin on the face and the palm of the hand detected whether the individual had high blood pressure. The tool could work just as well as a blood pressure cuff and also accurately detected type 1 or type 2 diabetes, according to a news release.¹

Artificial intelligence image with a computer | Image credit: © Song_about_summer | stock.adobe.com

The system, currently in early development in Japan, could minimize the need for blood tests ,blood pressure cuffs, or expensive wearable devices, according to researchers. The tool functions by analyzing subtle alterations in blood flow in the face and hands, caused by blood pressure and diabetes.¹

“This method may someday allow people to monitor their own health at home and could lead to early detection and treatment of high blood pressure and diabetes in people who avoid medical exams and blood tests,” Ryoko Uchida, BSc Pharm, a project researcher in the department of advanced cardiology at the University of Tokyo, said in a news release.¹

Pharmacists play a critical and growing role in screening for conditions such as hypertension and diabetes, potentially advancing hypertension outcomes across large populations. A more holistic approach to this care would involve interdisciplinary work, with interventions that address both pharmacological and nonpharmacological aspects of hypertension management. Pharmacists can also implement personalized care approaches and take into account patient-specific factors such as demographics, comorbidities, and lifestyle habits.² Having a simpler, faster way to screen for blood pressure and diabetes could further enable pharmacists’ involvement in this care.

In one study, researchers found that involvement of a community pharmacist in measuring blood pressure helped identify patients with undiagnosed hypertension, enabling the provision of medical care and treatment initiation. Pharmacists’ expertise in associated risk factors, drug therapies, and secondary lifestyle changes allows them to directly engage with patients on this issue.²

In the study, investigators tested the effectiveness of a high-speed video camera in capturing face and palm recordings at 150 images per second. Then, using wavelength data to detect pulse waves, they used an AI algorithm to detect high blood pressure and diabetes based on blood flow features in the skin.¹

According to the findings, compared with using blood pressure values measured by a continuous blood pressure monitor at the same time during video recording, the AI algorithm was 94% accurate in detecting stage 1 hypertension, defined according to the AHA’s guidelines for blood pressure 130/80 mm Hg or higher. Additionally, a 30-second video imaging/algorithm combination in a subset of patients was 86% accurate in detecting whether blood pressure was above normal, whereas a 5-second video was 81% accurate.¹

Similarly, compared with using hemoglobin A1c blood test results to screen for diabetes, the video and algorithm combination was 75% accurate in identifying individuals with diabetes.¹

“I was surprised about the applicability of the blood flow algorithm to detect diabetes,” Uchida said in the news release. “However, some of the major complications of diabetes are peripheral neuropathy—weakness, pain, and numbness, usually in the hands and feet—and other diseases related to blood vessel damage. It makes sense that changes in blood flow would be a hallmark of diabetes.”¹

Importantly, the study authors emphasized that several steps must be taken before this method could be ready for use outside of a research setting, such as incorporating an algorithm that considers arrhythmias. Future iterations could utilize an affordable sensor that only uses essential wavelengths and requires just a few seconds to gather data.¹

“Once it reaches that stage, it may be added to smartphones or even hung on a mirror where someone sits for a few moments, [and] may be mass-produced and inexpensive,” Uchida said.¹

AI-based tools are increasingly recognized for their great potential in identifying conditions such as diabetes and hypertension. Approaches like voice-based screenings would be extremely accessible via smartphone and have shown great accuracy.³

In one study, investigators asked 267 participants with and without type 2 diabetes to record 6- to 10-second clips of their voice up to 6 times a day for 2 weeks, collecting more than 18,000 voice samples. Using data on the differences in pitch, intensity, vibration, and breathiness or hoarseness, the team created an AI model that could detect diabetes with 89% accuracy in women and 86% accuracy in men. Importantly, the AI was able to flag voice differences even in individuals with similar age, sex, and body mass index.³

“It is really exciting to see more research that identifies ways to diagnose high blood pressure and diabetes non-invasively, 2 major risk factors for cardiovascular disease,” Eugene Yang, MD, MS, clinical professor of medicine in the cardiology division at the University of Washington School of Medicine in Seattle, said in a news release. “While the results are promising, it is important to recognize the validation of these technologies is lacking.”¹

Once the accuracy of diabetes detection is improved in their video-and-AI device, Uchida said the team hopes to seek FDA approval for an at-home device to detect diabetes.¹

“Currently, the only way to confirm the diagnosis of diabetes is invasive blood tests,” he said. “However, if it were to require only a non-invasive photo or video, that could be a game-changer.”¹

REFERENCES
1. AI-powered tool may offer quick, no-contact blood pressure and diabetes screening. News release. American Heart Association. November 11, 2024. Accessed November 15, 2024. https://newsroom.heart.org/news/ai-powered-tool-may-offer-quick-no-contact-blood-pressure-and-diabetes-screening-american-heart-association-scientific-sessions-2024-abstract-mdp1049?preview=070d&preview_mode=True

2. Waszyk-Nowaczyk M, Jasinska-Stroschein M, Dymek J, et al. The critical role of community pharmacists in blood pressure monitoring. Med Sci Monit. 2024;30:e944657-1-e944657-8. doi:10.12659/MSM.944657

3. Canning K. How AI Uses Voice to Detect Diabetes. Medscape. November 9, 2023. Accessed November 15, 2024. https://www.medscape.com/viewarticle/998280#:~:text=%22A%20voice%20screening%20tool%20could,body%20mass%20index%20(BMI)