AI-Powered Wearables Provide Breakthrough in IBD Disease Monitoring

Mount Sinai researchers demonstrate how wearable technology and artificial intelligence revolutionize inflammatory bowel disease management, offering unprecedented early detection and personalized care.

Research: Physiological Data Collected from Wearable Devices Identify and Predict Inflammatory Bowel Disease Flares. Image Credit: BAZA Production / Shutterstock

Wearable devices can identify, differentiate, and predict flare-ups, or the worsening of symptoms and inflammation, in inflammatory bowel disease (IBD), Mount Sinai researchers have shown in a first-of-its-kind study.

The findings, published in the journal Gastroenterology on January 16, suggest that wearable technology can predict the subsequent development of flares in IBD, enabling continuous disease monitoring through widely available commercial devices.

"Current disease-monitoring methods rely on patients directly interacting with their doctors, either through office visits, blood or stool testing, or by undergoing a colonoscopy. These methods also only assess the disease at one point in time, and can often be invasive or inconvenient," said first author Robert Hirten, MD, Clinical Director of the Hasso Plattner Institute for Digital Health; and Associate Professor of Medicine (Gastroenterology), and Artificial Intelligence and Human Health, at the Icahn School of Medicine at Mount Sinai. "Our study shows that commonly used wearable devices such as Apple Watches, Fitbits, and Oura Rings can be effective tools in monitoring chronic inflammatory diseases like IBD. This creates an opportunity to monitor the disease remotely outside the health care setting, in a continuous manner, and potentially in real time."

IBD is a chronic condition that causes inflammation in the intestines and affects more than 2.4 million people in the United States. Mount Sinai researchers enrolled more than 300 participants with ulcerative colitis or Crohn's disease, the two major types of IBD, from 36 states. The participants wore devices, answered daily symptom surveys, and provided blood and stool assessments of inflammation. These biological assessments, including C-reactive protein and fecal calprotectin levels, were used to validate wearable device data.

The researchers found that circadian patterns of heart rate variability (HRV), a key indicator of autonomic nervous system function, along with heart rate, oxygenation, and daily activity, all measured by the wearable devices, were significantly altered when inflammation or symptoms were present. Moreover, these physiological markers could detect inflammation even in the absence of symptoms and distinguish whether symptoms were driven by active inflammation in the intestines. Specifically, HRV, heart rate, and resting heart rate (RHR) metrics were used to differentiate between inflammatory flares and symptomatic but non-inflammatory episodes. Importantly, the researchers found that these metrics measured by wearables changed up to seven weeks before flares developed. Statistical techniques, such as linear mixed-effect models, were employed to assess longitudinal data and provide robust predictive insights.

The researchers are applying similar approaches to other chronic inflammatory diseases, such as rheumatoid arthritis, and leveraging artificial intelligence to develop algorithms using wearable device data to predict flares on an individualized basis. This integration of AI enhances the ability to provide precise, personalized monitoring for patients. "These findings open the door to leveraging wearable technology for health monitoring and disease management in innovative ways we haven't previously considered," Dr. Hirten said. "Our hope is that, in the future, this approach will significantly enhance the quality of life of our patients."