July 15, 2024 — HeartBeam, Inc., a medical know-how firm centered on reworking cardiac care by the ability of customized insights, at this time introduced new examine information demonstrating that HeartBeam AI mixed with vectorcardiography (VCG) outperformed an skilled panel of coronary heart rhythm cardiologists in detecting atrial flutter. HeartBeam AI is the corporate’s deep studying (a type of AI) algorithm for detecting abnormalities in the timing or sample of heartbeats. The info was offered by Joshua M. Lampert, MD, Cardiac Electrophysiologist, Assistant Professor of Drugs, and Medical Director of Machine Studying for Mount Sinai Fuster Heart Hospital on the Icahn Faculty of Drugs at Mount Sinai, throughout the Heart Rhythm Society annual assembly in Boston.
Within the examine, HeartBeam AI was utilized to a set of 173 VCGs, single-lead ECGs and 12-lead ECGs to determine atrial flutter. The identical set of single-lead ECGs and 12-lead ECGs was reviewed by a panel of three electrophysiologists (EP panel) for atrial flutter, a typical arrhythmia that considerably will increase a affected person’s threat for stroke. Key findings from the evaluation present that HeartBeam AI mixed with VCG:
- Outperformed an skilled panel reviewing single-lead ECGs, with a statistically important 40% enchancment in the detection of atrial flutter instances (sensitivity: 97.3% for HeartBeam AI+VCG vs. 69.4% for EP panel).
- Demonstrated a statistically important 6% enchancment in the detection of atrial flutter instances in comparison with an skilled panel reviewing 12-lead ECGs (sensitivity: 97.3% for HeartBeam AI+VCG vs. 91.1% for EP panel).
- Delivered zero variability in the detection of atrial flutter in comparison with the EP panel.
Further particulars concerning the examine will be discovered here.
Dr. Lampert commented, “The trendy vectorcardiogram is sort of 100 years outdated, and but we’re cautiously respiration new life into it with the arrival of novel acquisition applied sciences and deep studying algorithms. This examine demonstrates {that a} deep studying algorithm utilized to a reworked VCG performs comparably as nicely when utilized to the gold-standard 12-lead ECG. The AI algorithm general outperformed a panel of electrophysiologists in distinguishing atrial flutter from sinus rhythm with excellent settlement between a number of mannequin predictions in comparison with important interobserver variability amongst electrophysiologists, a discovering significantly notable on single lead ECG evaluation.”
HeartBeam’s core vectorelectrocardiography (3D VECG) know-how captures the guts’s indicators in three projections (X, Y, Z), much like VCG, and synthesizes a 12-lead ECG. The Firm’s first deliberate utility of the 3D VECG platform know-how is HeartBeam AIMIGo™, a credit score card-sized machine for affected person use at house or wherever, which is presently beneath assessment with FDA.
By leveraging AI to research the data-rich indicators, HeartBeam believes will probably be capable of enhance diagnostic accuracy and to unlock a extra customized method to cardiac take care of sufferers. As a affected person makes use of AIMIGo over time, there might be a collection of ECG readings. HeartBeam goals to leverage AI to research the information to offer a longitudinal view of the affected person’s cardiac standing and transfer past at this time’s 12-lead ECGs, which generally solely present a snapshot in time.
“The info is extremely encouraging, showcasing the potential of our synthetic intelligence program to enhance diagnostic accuracy when a affected person is exterior of a medical facility,” stated Branislav Vajdic, PhD, CEO and Founding father of HeartBeam. “We’ll proceed to construct upon this robust basis as we advance our AI program to revolutionize cardiac care administration in the longer term.”
Dr. Lampert has no related conflicts of curiosity and HeartBeam didn’t fund his participation in this work or make the choice to submit the evaluation for presentation.
For extra info: www.heartbeam.com