Can blood pressure be screened using images of the face?

Researchers explored whether an artificial intelligence-based approach can improve the early detection of hypertension.

In a study published in Scientific Reports, the researchers developed a noninvasive, camera-based method that used deep-learning technology. They then used facial images from just over 500 patients with and without hypertension to train the model to predict hypertension using images of the whole face or six anatomical regions.

The AI model exhibited a mean Intersection over Union of 98.4% using segmented facial regions and an accuracy of 83% using whole facial images. The zygomatic and buccal segments demonstrated comparable accuracy as the whole facial images — representing the potential for these segments to reveal physiologic changes linked to hypertension.

Because of the underdiagnosis of hypertension caused by the shortcomings of current real-world measurement strategies, the researchers indicated that their novel AI-based approach could enhance hypertension screenings, particularly in resource-limited medical settings.

Read more: Scientific Reports

The article presented here is intended to inform you about the broader media perspective on dentistry, regardless of its alignment with the ADA's stance. It is important to note that publication of an article does not imply the ADA's endorsement, agreement, or promotion of its content.