Hidden Hypertension: AI-Powered Detection in Health Records
- Boston—artificial intelligence may hold the key to unlocking hidden clues about hypertension within electronic health records (EHRs).
- Researchers used AI to analyze heart ultrasounds, identifying patients with heart muscle thickening, a common indicator of hypertension.
- The study involved 648 Mass General Brigham patients not previously diagnosed with hypertension.
AI is revolutionizing hypertension detection! A groundbreaking study reveals how artificial intelligence, using natural language processing, can identify hidden signs of this primary_keyword within electronic health records (EHRs).Doctors notified of potential risks saw diagnoses surge nearly fourfold after AI flagged crucial indicators like heart muscle thickening. This innovative approach offers a proactive strategy to elevate heart care. The study, published in JAMA Cardiology, highlights the power of leveraging existing data, transforming routine care through early intervention. News Directory 3 recognizes how this technology can improve health outcomes. Explore the future of proactive healthcare: Discover what’s next for secondary_keyword management and broader implementation.
AI identifies Hypertension Risk via Electronic Health Records
Updated June 4, 2025
Boston—artificial intelligence may hold the key to unlocking hidden clues about hypertension within electronic health records (EHRs). A Mass General Brigham study, published in JAMA Cardiology, reveals how natural language processing can pinpoint patients at risk. The study was also presented at the American College of Cardiology’s Annual Scientific Session & Expo.
Researchers used AI to analyze heart ultrasounds, identifying patients with heart muscle thickening, a common indicator of hypertension. When doctors were alerted to these findings, new hypertension diagnoses rose nearly fourfold, along with prescriptions for blood pressure medication.
The study involved 648 Mass General Brigham patients not previously diagnosed with hypertension. The average age was 59, and 38% were women. Half received the AI-driven intervention, where a coordinator informed their physicians of the potential risk and offered resources like blood pressure monitoring and cardiology evaluations. The control group received standard care.
Adam Berman, MD, MPH, lead author and assistant professor at NYU Grossman School of Medicine, noted the vast amount of data generated during routine care. “There are frequently enough subtle clues…that may indicate a patient has hypertension,” Berman said. “the premise of our trial was that the data are likely hiding in plain sight, and we wanted to validate methods of bringing it to light to improve the care of our patients.”
Jason H. Wasfy, MD, MPhil, senior author from Massachusetts General Hospital, emphasized the importance of targeted interventions. “Clinicians are often overloaded with alerts that can cause fatigue and burnout, so we intentionally designed our outreach to be delivered by a person,” Wasfy said.
The intervention group saw a notable increase in new hypertension diagnoses (15.6% vs. 4.0%) and prescriptions for antihypertensive medication (16.3% vs. 5.0%) compared to the control group. Physician response to the alerts was largely positive, with 72% reacting favorably.
What’s next
Future research will explore automating the notification process for broader implementation while maintaining its effectiveness in diverse healthcare settings. The goal is to leverage existing data to improve healthcare delivery and patient outcomes in hypertension management.