AI-Enabled ECG Helps Determine Future Risk of Cognitive Decline and Stroke

One-third of ischemic strokes, the most common type of stroke, are caused by atrial fibrillation, the most familiar cardiac rhythm abnormality. However, since many patients are asymptomatic, atrial fibrillation is underdiagnosed.

In a population-based study conducted by researchers at Mayo Clinic, artificial intelligence-enabled electrocardiography (ECG) was recently shown to detect the presence of brief episodes of atrial fibrillation, and the competence of an AI-enabled ECG algorithm to foretell atrial fibrillation up to 10 years before clinical diagnosis was confirmed.

A new Mayo Clinic population-based research now shows that the algorithm can help pinpoint patients who are at higher risk of cognitive decline. According to the research, an AI-enabled ECG with a high probability of atrial fibrillation was also linked to the presence of infarctions, or incidents of cerebral stroke, on MRI.

The observations were published in the Mayo Clinic Proceedings.

The majority of the infarctions were subcortical — they occur beneath the cortex of the brain. This implies that an AI-enabled ECG not only identifies other cardiac disease markers and coincides with small vessel cerebrovascular disease and cognitive decline, but also predicts atrial fibrillation.

This study finds that artificial intelligence-enabled electrocardiography acquired during normal sinus rhythm was associated with worse baseline cognition and gradual decline in global cognition and attention.

Jonathan Graff-Radford MD, Study Corresponding Author and Neurologist, Mayo Clinic

“The findings raise the question whether initiation of anticoagulation is an effective and safe preventive strategy in individuals with a high AI-ECG algorithm score for reducing the risk of stroke and cognitive decline,” adds Jonathan Graff-Radford.

According to Dr. Graff-Radford, prospective controlled investigations are necessary to see if a high atrial fibrillation score on an AI-enabled electrocardiogram can be used as a biomarker to identify patients who need anticoagulation or more aggressive stroke risk factor modification.

The retrospective research reviewed sinus-rhythm ECGs of 3,729 patients with a median age of 74 years who were enrolled in the Mayo Clinic Study of Aging between 2004 and 2020. When demographic factors were taken into account, the AI-enabled ECG atrial fibrillation score was linked to a lower baseline and a faster decline in global cognitive scores.

A high atrial fibrillation probability in the ECG was associated with MRI-detected cerebral infarcts in about one-third of the patients who had an ECG and an MRI.

Application of this AI-ECG algorithm may be another way to screen individuals not only to determine risk of atrial fibrillation, but also to identify future risk of cognitive decline and stroke.

Jonathan Graff-Radford MD, Study Corresponding Author and Neurologist, Mayo Clinic

The National Institute on Aging and the National Institutes of Health funded the research. The Rochester Epidemiology Project made the research possible. The article identifies potential competing interests. Among the potential competing interests, Peter Noseworthy MD, a Mayo Clinic cardiologist, and Mayo Clinic have filed patents related to the application of AI to ECG for diagnosis and risk stratification.

AI-Enabled ECG for AF Identifies Cognitive Decline Risk and Cerebral Infarcts. Video Credit: Mayo Clinic.

Journal Reference:

Weil, E. L., et al. (2022) Artificial Intelligence—Enabled Electrocardiogram for Atrial Fibrillation Identifies Cognitive Decline Risk and Cerebral Infarcts. Mayo Clinic Proceedings. doi.org/10.1016/j.mayocp.2022.01.026.