7 August 2019

AI may detect ‘invisible’ rhythm on ECGs

Cardio Research Technology

Artificial intelligence (AI) may be the future of diagnosing atrial fibrillation and could one day detect high-risk patients from the convenience of the consulting room, an early stage study suggests.

A US article, recently published in The Lancet, found an AI system to be 83% effective in detecting intermittent or impending atrial fibrillation even when patients were in sinus rhythm at the time of their ECG.

The study authors trained a neural network by creating a pattern recognition program from the heart rhythms of more than 180,000 patients.

The study included patients (from 18 years old) who had recorded at least one normal sinus rhythm (standard 10-second 12-lead ECG) between 1993 and 2017 at the Mayo Clinic laboratory in Minnesota.

“We found that an AI model can differentiate between patients with a history of (or impending) atrial fibrillation with a high degree of accuracy using a single routine ECG.

“Addition of multiple ECGs within an individual patient improved the model accuracy and suggests repeated measures might yield even better performance,” the study authors said.

The authors conceded, however, that because the AI was trained from the ECGs of patients who needed clinical investigations, further investigations would be needed to determine the effectiveness of the technology on the general population.

Intermittent atrial fibrillation is frequently asymptomatic and often goes undiagnosed in patient and existing screening methods often require costly and prolonged monitoring.

“We aimed to develop a rapid, inexpensive, point-of-care means of identifying patients with atrial fibrillation using machine learning,” the study authors said.

Dr Jeroen Henriks (PhD), from the Centre for Heart Rhythm Disorders at the University of Adelaide, said the study represented an innovative approach to develop an AI-enabled ECG.

“Given that AI algorithms have recently reached cardiologist level in diagnostic performance, this AI-ECG interpretation is ground-breaking in creating an algorithm to reveal the likelihood of atrial fibrillation in ECGs showing sinus rhythm,” he said.

In addition, the study authors said an AI-enabled ECG could allow point-of-care identification of patients who had a high risk of atrial fibrillation.

“This result could have important implications for atrial fibrillation screening and for the management of patients with unexplained stroke,” they concluded.

The Lancet 2019, 1 August