Text Your Heart to the Doctor

Remote cardiac ECGs accuracy improved

(RxWiki News) The technology already exists to send electrocardiogram (ECG) images via smart phone from a rural community to a hospital, potentially thousands of miles away. But errors still happen.

The creation of a new algorithm capable of analyzing ECG interference from misplaced or disturbed electrodes from patient movement or electromagnetic noise, designed to be operated on a smart phone is taking imaging high tech.

"Wireless technology can help rural patient populations reach doctors."

Xiaopeng Zhao, assistant professor in the department of mechanical, aerospace and biomedical engineering at the University of Tennessee, said there is a large population that does not receive adequate health care because they live in rural locations. 

The main goal in creating the technology was to reduce misdiagnoses and save lives, particularly in rural communities. It also can be helpful for novice health workers and busy intensive care units. He said the algorithm helps bring doctors to patient's homes through mobile technology capable of helping to reduce errors and improving ECG quality.

Though accurate, there is still room for error in taking ECGs, most commonly used as a screening tool for heart abnormalities. These mistakes can lead to misdiagnosis. It is estimated that about 4 percent of ECGs are taken with misplaced electrodes, which can lead to faulty diagnoses and treatment.

Zhao's algorithm is more reliable than conventional ones because it simultaneously looks for irregular patterns caused by interference. The algorithm then assigns a letter grade between A and F, designating any specific weaknesses in the test. The algorithm also recommends where to place electrodes for accuracy.

It runs on a java platform and can be installed and operated on a smart phone. It can access an ECG within 10 seconds since speed can be key in a life and death situation.

The algorithm recently won the top spots in the Physionet Challenge 2011, taking first, first and third places. The challenge asks researchers to propose solutions to problems that are unsolved or not well solved. A new topic is announced each year by sponsors Physionet and the National Institutes for Health.

The research will be presented at the annual Computing in Cardiology conference in China next month.

Reviewed by: 
Review Date: 
August 25, 2011