This New AI Algorithm Can Make ECG Diagnostic on Your Fitbit, Apple Watch More Accurate

This advancement improves the specificity and sensitivity of atrial fibrillation (AF) detection when compared to existing algorithms from each device manufacturer.

PulseAI, an artificial intelligence healthcare solutions provider, revealed a significant development in electrocardiogram (ECG) monitoring on consumer devices on Thursday. The new paper, "Reducing the Burden of Inconclusive Smart-Device Single-lead ECG Tracings via a Novel Artificial Intelligence Algorithm," published in the Cardiovascular Digital Health Journal, highlights the important function of PulseAI's neural network algorithm in cardiac monitoring.

A team of cardiologists and researchers from the University Hospital Basel in Switzerland conducted the investigation. The findings show that the neural network algorithm used by PulseAI dramatically minimises inconclusive diagnoses from commercially accessible consumer ECG devices like Apple, Samsung, AliveCor, and Fitbit.

This advancement improves the specificity and sensitivity of atrial fibrillation (AF) detection when compared to existing algorithms from each device manufacturer. Atrial Fibrillation is considered the foremost precursor to stroke.

Methodology Used in the Study

For the purpose of study, the team of researchers assessed the accuracy of smartwatches. They found a surprisingly high number of inconclusive ECGs, Dr Patrick Badertscher, principal investigator and cardiac electrophysiologist at the University Hospital Basel, Switzerland said.

"This study confirms that the use of AI solutions may be extended to smart-device ECGs rhythm classification. The proposed AI-based algorithm could be valuable to assist physicians with managing the large amount of data in the detection of AF using smart-devices,” Dr Badertscher added.

The latest discovery has potential to ease the clinical burden due to the manual review of smart-device ECGs.

Northern Ireland-based PulseAI is a medical technology company that develops artificial intelligence-enabled software technologies to increase the accuracy, efficiency, and scalability of cardiac diagnostics. PulseAI has designed a cloud-based platform that allows for easy ECG device connection. The platform also enables scalable deployment of patented deep learning algorithms based on a growing library of more than 6 million ECGs from 2 million patients.

Dr Alan Kennedy, CEO of PulseAI believes that the fresh development marks a new era in cardiac healthcare. "Our AI algorithm's ability to provide clear, accurate diagnostic readings from consumer devices will dramatically improve how we monitor and diagnose heart conditions at scale, making cardiac diagnostics more efficient and accessible for patients worldwide."

Artificial Intelligence In ECG

An electrocardiogram (ECG) is a key diagnostic tool used to assess the cardiovascular system's performance. It measures electric signals and heartbeat of the patient to diagnose and treat various cardiac problems, making it a vital tool in modern medicine.

While automated ECG interpretation has been widely used, a number of studies have shown that computer-aided ECG reading is still generally erroneous to this day.

However, Artificial Intelligence is fast changing it. AI-powered ECG interpretation based on machine learning algorithms and deep neural networks are turning out to be as accurate as an expert cardiologist with 30 years of expertise.

Recently, a team of researchers at Mount Sinai's Icahn School of Medicine developed an artificial intelligence tool to assist forecast which cardiovascular patients are at increased risk of having poor right ventricular function. A deep learning-ECG model was trained by the study team to predict right ventricular dilatation, dysfunction, and numerical RVEDV and RVEF.

In August last year, teams from Brigham and Women's Hospital in Boston and Keio University in Japan developed a deep learning model that could accurately screen ECGs for signs of atrial septal defects. ASD is a type of congenital heart defect that can harm the cardiovascular system and result in life-threatening complications if not treated promptly.

Anumana, another AI-driven HealthTech company last year received US FDA breakthrough device designation for its artificial intelligence-guided ECG algorithm for the early identification of cardiac amyloidosis.