A machine-learning algorithm can accurately screen for hypertension and diabetes using brief facial video recordings, offering a contactless diagnostic option.
Key Details
- 1Researchers from the University of Tokyo developed an AI algorithm to analyze facial and palm video recordings for undiagnosed hypertension and diabetes.
- 2In a prospective study of 215 participants, the algorithm detected hypertension with over 90% accuracy from a 5-second facial video and up to 95% with longer recordings.
- 3Diabetes was identified with 81.2% accuracy from 5-second videos and 88.2% accuracy from 30-second videos.
- 4The AI could estimate systolic blood pressure, achieving a mean error within the AAMI clinical criterion for bias but with higher-than-acceptable standard deviation.
- 5This contactless technology may allow mass public screening without traditional cuffs or blood sampling, pending further validation in larger, more diverse populations.
Why It Matters
This research demonstrates the expanding capabilities of imaging AI in non-traditional modalities (facial video) for scalable, contactless medical screening. If validated, it could significantly increase early detection and intervention for hypertension and diabetes, impacting public health and preventive care strategies.
Source
EurekAlert
Related News
•EurekAlert
NYU AI Model Improves 5-Year Breast Cancer Risk Prediction via 3D Mammograms
NYU researchers developed an AI model using longitudinal 3D mammograms that outperforms single-scan and 2D-based tools in predicting 5-year breast cancer risk.
•EurekAlert
AI Framework Accelerates Aortic Aneurysm Risk Prediction from Imaging
Researchers developed BioPINN-LM, combining physics-informed neural networks and multimodal large language models to deliver fast, interpretable risk assessments for ascending thoracic aortic aneurysms.
•EurekAlert
Study Finds Patient Voices Missing in Generative AI Design for Oncology
A Flinders University-led review found patients and carers are rarely involved in shaping generative AI tools used in oncology.