An Untethered Heart Rhythm Monitoring System with Automated AI-Based Arrhythmia Detection for Closed-Loop Experimental Application
Shanliang Deng (Leiden University Medical Center, TU Delft - Electronic Components, Technology and Materials)
B.L. den Ouden (Leiden University Medical Center, TU Delft - Electronic Components, Technology and Materials)
Tim De Coster (Leiden University Medical Center)
Cindy I. Bart (Leiden University Medical Center)
Wilhelmina H. Bax (Leiden University Medical Center)
René H. Poelma (TU Delft - Electronic Components, Technology and Materials)
Antoine A.F. de Vries (Leiden University Medical Center)
Kouchi Zhang (TU Delft - Electronic Components, Technology and Materials)
Vincent Portero (Leiden University Medical Center)
D.A. Pijnappels (Leiden University Medical Center)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
The heart produces bioelectrical signals, which can be measured as an electrocardiogram (ECG) for the detection of rhythm disturbances. Rapid and precise detection of these arrhythmias is crucial for their termination by closed-looped therapeutic interventions to counteract detrimental effects. However, there is a current lack of such systems tailored for experimental cardiovascular applications. This hampers not only in-depth mechanistic studies but also translational testing of new therapeutic strategies, especially in an untethered manner in awake animal models. To break new ground, recent advances to develop a non-invasive AI-supported heart rhythm monitoring system for untethered automated arrhythmia detection in a continuous manner is combined. This system is housed in a lightweight jacket for mobile use and includes an on-skin ECG sensor, a low-power microprocessor unit, a massive data storage unit, and a power-management system. By implementing a novel hybrid algorithm based on so-called heart rate (R-R) variability and a case-specific AI model, 100% sensitivity and 95% specificity is achieved in detecting atrial arrhythmias within 2 s upon initiation in adult rats. Thereby, the novel system sets the stage for advanced mechanistic studies and therapeutic testing, including closed-loop applications aiming for the termination of a broad range of atrial arrhythmias.
help_outline