SBIR Phase I: Real-time arrhythmia detection and classification using a waterproof armband

Mobile Sense Technologies, Inc., Farmington CT

Investigators

Abstract

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to help identify people that have paroxysmal or asymptomatic cardiac arrhythmias before they become dangerous. Atrial Fibrillation (AF) is the most common arrhythmia, and the prevalence of AF increases with age. AF has a profound impact on longevity and quality of life. Patients typically develop paroxysmal, short-lived episodes of the arrhythmia and are frequently asymptomatic. This often leads to delays in diagnosis of AF until later stages when the arrhythmia is more persistent. Even short episodes of paroxysmal AF are associated with increased risk for stroke, heart failure, hospitalization, and death. The population with undiagnosed AF is substantial and studies have shown that continuous ECG monitoring for more than 3 years is required to achieve close to 100% AF detection. Current 30-day monitors detect 5% of paroxysmal and asymptomatic cases, since events must occur while the monitor is being worn. An armband-based approach offers a way to overcome the current restrictions in chest based monitors and bridges the gap between short-duration 30-day monitors and long-term implanted monitors. The proposed project uses an armband-based device to capture electrocardiogram (ECG) and electromyogram (EMG) signals on the upper arm using a waterproof and wireless device. The sensors are made from a composite material that can operate both wet and dry and does not require adhesives or hydrogels to capture all aspects of the waveforms. The challenge with the upper arm is that the ECG signals have less magnitude then over-the-heart and it is necessary to remove motion / noise artifacts as well as EMG signals to create a clean ECG signal. The research is focused on embedding algorithms into the device and testing and refining them for removing motion and noise artifacts that are captured by an accelerometer and to remove EMG signals. After a clean signal is established, proven algorithms for classification of the arrhythmias can be run embedded in the device.