A new study by researchers at Mass General Brigham reveals that artificial intelligence (AI) can detect hidden cases of hypertension by analyzing electronic health records (EHR). Using natural language processing, the team identified patients with thickened heart muscle—a sign of untreated high blood pressure—and found that notifying physicians led to nearly four times more diagnoses and prescriptions for hypertension. The findings, published in JAMA Cardiology, were also presented at the 2025 American College of Cardiology’s Annual Scientific Session & Expo.

Hypertension, often called the "silent killer," frequently goes undiagnosed because it lacks noticeable symptoms. In the U.S., nearly half of those affected are unaware of or untreated for the condition. The study aimed to address this gap by leveraging existing EHR data.

The research team developed an AI algorithm to scan echocardiogram reports for signs of left ventricular hypertrophy, a heart condition commonly linked to high blood pressure. They identified 648 patients at Mass General Brigham who had no prior diagnosis of heart muscle issues or hypertension. Half of these patients were randomly selected for an intervention where their doctors’ received notifications about the findings, along with resources for further evaluation. The other half received standard care.

Patients whose doctors were notified were significantly more likely to be diagnosed with hypertension (15.6% vs. 4.0%) and prescribed medication (16.3% vs. 5.0%) compared to the control group. Clinicians responded positively to the intervention, with 72% expressing favorable feedback.

The study highlights the potential of AI to enhance healthcare by uncovering critical information buried in routine clinical data. While the current approach involved human coordinators to avoid alert fatigue, the team is exploring ways to scale the method for broader use.

“This isn’t about replacing traditional care but augmenting it,” said lead author Dr. Adam Berman. “We’re using data that’s already there to improve patient outcomes.”

The research underscores how innovative technologies can transform patient care by turning overlooked data into actionable insights.