Separation of ventricular tachycardia from sinus rhythm using a practical, real-time template matching computer...

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Implantable cardioverter-defibrillators (ICDs) require accurate detection of ventricular tachyarrhythmias.
• Distinguishing ventricular tachycardia (VT) from sinus tachycardia (ST) or supraventricular tachycardia (SVT) is challenging due to rate overlap.
• Template matching morphology analysis offers a potential solution to reduce false detections.

Purpose of the Study:

• To evaluate the real-time accuracy of intraventricular electrogram (IVEG) template matching for separating VT from sinus rhythm (SR) within ICD constraints.
• To compare the performance of amplitude normalized area of difference (NAD) with signature analysis (SIG) for VT detection.

Main Methods:

• Digitization of IVEGs from 16 patients undergoing electrophysiological study (EPS) at 250 Hz with 6-bit quantization.
• Filtering (16-44 Hz) and differentiation of IVEG signals.
• Real-time template matching using NAD and SIG algorithms, with SR templates selected and updated periodically.

Main Results:

• Template matching analysis demonstrated accuracy in separating VT from SR electrograms on an ICD sensing emulator.
• Signature analysis (SIG) was evaluated as a computationally less expensive alternative to NAD.
• Eleven-beat running medians of beat-by-beat NAD and SIG results were determined for performance assessment.

Conclusions:

• Real-time IVEG template matching is a viable method for differentiating VT from SR in ICDs.
• Signature analysis (SIG) presents a computationally efficient approach for VT detection, potentially suitable for ICD implementation.
• Further validation is needed to confirm the real-world performance of these algorithms under various clinical conditions.