ECG Analysis Platform Detects High Risk AF Patients

A novel system uses 12-lead electrocardiogram (ECG) data to identify patients at high risk of developing atrial fibrillation (AF) within the year.

Developed by Tempus (Chicago, IL, USA), in collaboration with the Geisinger Health System (Geisinger; Danville, PA, USA), the Tempus ECG Analysis Platform evaluates the results of a 12-lead ECG exam administered as part of routine care in order to provide clinicians with insight into a patient's risk of future AF and/or atrial flutter events. The deep neural network algorithm was trained using 1.6 million ECGs from 430,000 patients over 35 years of patient care (among patients without a previous history of AF), who would develop it within 12 months.


In a clinical study held at Geisinger, which was published on February 16, 2021 in Circulation, Tempus ECG Analysis performance exceeded that of current clinical models for predicting AF risk. In addition, 62% of patients without known AF, and who experienced an AF-related stroke within three years were identified as high risk by the model before the stroke occurred. When interpreted in conjunction with other clinical information, the platform could help clinicians pursue early and proactive diagnoses for improved clinical management of these conditions and their associated health risks.

“Every year, hundreds of millions of routine ECGs are performed in the United States to detect cardiac abnormalities,” said Joel Dudley, PhD, chief scientific officer of Tempus. “We are making ECGs smarter so that they can identify the risk of future, highly treatable clinical events of interest, such as atrial fibrillation, thus enabling clinicians to act earlier in the course of disease and improve patient outcomes.”

“Not only can we now predict who is at risk of developing atrial fibrillation, but this work shows that the high risk prediction precedes many AF-related strokes,” said Brandon Fornwalt, MD, PhD, co-senior author of the study and chair of Geisinger's Department of Translational Data Science and Informatics. “With that kind of information, we can change the way these patients are screened and treated, potentially preventing such severe outcomes. This is huge for patients.”

In patients with AF, rapid, disorganized electrical signals cause the atria to quiver, interrupting the normal synchronized rhythm between the atria and the ventricles. As a result, the ventricles may beat fast and without a regular rhythm, leading to blood clots, strokes, heart failure, and other heart-related complications.

Related Links:
Tempus
Geisinger Health System