Artificial Intelligence Can Predict Worsening of Heart Failure Before Hospitalization

Newly developed wearable sensor based on Artificial Intelligence (AI) technology can remotely detect any deterioration in the condition of heart failure patients, days ahead of a possible health crisis thereby reducing the need hospitalization, reports a new study conducted by physician-scientists at the University of Utah School of Medicine and Baylor College of Medicine, USA.

The research team has indicated that this novel device has the potential to prevent 1 in 3 hospital readmissions of heart failure patients, following initial discharge from the hospital. Moreover, continuous monitoring of the condition of heart failure patients using the new device will enable them to lead a better quality of life. The research findings have been published in Circulation: Heart Failure, a constituent journal of the American Heart Association (AHA), Dallas, Texas, USA.

Advertisement

Study Team

The study was led by Dr. Josef Stehlik, MD, MPH, who is the Christi T. Smith Professor of Medicine at the University of Utah School of Medicine, Salt Lake City, Utah. He is also the Medical Director of the Heart Transplant Program and Co-Chief of the Advance Heart Failure Program at the University of Utah Hospital and the George E. Wahlen Department of Veterans Affairs Medical Center, Salt Lake City, Utah, USA.

Dr. Biykem Bozkurt, MD, PhD was a study collaborator and co-author of the paper. She is a Professor of Medicine-Cardiology and the Mary and Gordon Cain Chair of Internal Medicine at the Winters Center for Heart Failure Research, Baylor College of Medicine, Houston, Texas, USA.

Advertisement

Heart Failure and its Complications

Heart failure, also known as congestive heart failure, is a chronic condition in which the heart is not able to pump blood effectively. This arises from the narrowing of the coronary arteries due to the deposition of cholesterol (atherosclerosis), which gives rise to coronary artery disease (CAD). High blood pressure (hypertension) can worsen the condition.

In the US, approximately 6.2 million people are currently living with heart failure, which tops the list of diagnosed conditions at the time of discharge from the hospital. Over 30 percent of these patients are likely to be readmitted within 90 days of discharge due to the following complications:

• Breathing difficulty
• Coughing or wheezing
• Fatigue and weakness
• Rapid heartbeat (tachycardia)
• Irregular heartbeat (arrhythmia)
• Fluid build-up (edema) in the legs and feet

Moreover, in many instances, hospitalization can reduce the patient’s ability to independently take care of themselves.

“Those individuals who have repeated hospitalizations for heart failure have significantly higher mortality,” says Bozkurt. “Even if patients survive, they have poor functional capacity, poor exercise tolerance and low quality of life after hospitalizations. This patch is a new diagnostic tool, which could potentially help us prevent hospitalizations and decline inpatient status.”

Advertisement

Salient Features of the Study

The key features of the study are indicated below:

• 100 heart failure patients were included in the study
• Average age of the patients was 68 years
• The patients were treated at Veterans Affairs hospitals located in the following 4 cities:
• Salt Lake City, Utah
• Houston, Texas
• Palo Alto, California
• Gainesville, Florida
• Following discharge, the patients were given an adhesive sensor patch
• The patients wore the sensor patches on their chests, 24 hours a day for 3 months
• The sensors were manufactured by PhyslQ – an IT company based in Chicago that produces sensors for monitoring patients
• The sensors monitored the following major parameters:
• Heart rate and heart rhythm
• Respiratory rate
• Walking or running
• Sleep patterns
• Body posture
• These AI-based sensors collected normal baseline data of the patients
• Bluetooth transmitted the data to a smartphone that relayed it to an analytics platform connected to a secure server
• Deviation of the data from normal values alerted doctors that the patient’s condition was deteriorating
• The AI technology accurately predicted the need for hospitalization more than 80 percent of the time
• The alert was sent to the doctors on an average 6.5 days before the patient was readmitted

“This study shows that we can accurately predict the likelihood of hospitalization for heart failure deterioration well before doctors and patients know that something is wrong,” says Stehlik. “Being able to readily detect changes in the heart sufficiently early will allow physicians to initiate prompt interventions that could prevent rehospitalization and stave off worsening heart failure.”

Future Plans

The research team is planning to conduct a large-scale clinical trial of the AI-based sensors to assess the following:

• Ability of the sensors to generate statistically significant results in a large cohort of patients for validating the results of the present study that used a small sample size (n=100)
• Effectiveness of the sensors to instantly alert doctors for instituting early interventions to reduce the number of hospital readmissions of heart failure patients

Concluding Remarks

“There’s a high risk for readmission in the 90 days after initial discharge,” says Stehlik. “If we can decrease this readmission rate through monitoring and early intervention, that’s a big advance. We’re hoping even in patients who might be readmitted that their stays are shorter, and the overall quality of their lives will be better with the help of this technology.”

Funding Source

The research was funded by the Department of Veterans Affairs Office of Information & Technology and the Veterans Health Administration (VHA) Innovation Ecosystem.

Reference:

1. Continuous Wearable Monitoring Analytics Predict Heart Failure Hospitalization - (https://www.ahajournals.org/doi/10.1161/CIRCHEARTFAILURE.119.006513)

Source-Medindia