Cardiac Event Detection Using Fourth-Order Difference Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Implantable cardioverter defibrillators face challenges in accurately distinguishing between supraventricular tachycardia (SVT) and ventricular tachycardia (VT) due to similar tachycardia cycle lengths, leading to inappropriate therapy delivery and potential delays in life-saving interventions.
Innovation Solution
An apparatus and method for cardiac event detection using improved techniques to generate and align cardiac cycle templates with unknown heart rhythms, employing fourth-order difference signals and normalized waveform area difference (NWAD) to classify beats as SVT or VT, enabling precise morphology matching and therapy decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If rate-based detection is used to detect fast ventricular rate, then tachycardia detection speed is improved, but accuracy in distinguishing SVT from VT deteriorates
Solution Approach 1:
The detection process is segmented into two independent stages: rate-based detection for rapid tachycardia identification, followed by morphology analysis for precise SVT/VT differentiation. This segmentation allows the system to benefit from both fast rate detection and accurate morphology-based classification without compromise.
Solution Approach 2:
Waveform morphology analysis serves as an intermediary step between rapid rate detection and therapy delivery decisions. The morphology comparison acts as a mediator that refines the initial rate-based detection result, enabling accurate SVT/VT discrimination while maintaining rapid response capability.
2Reliability
If high-voltage cardioversion shocks are delivered to terminate tachycardia, then therapy effectiveness is improved, but patient comfort and battery consumption worsen
Solution Approach 1:
The system performs preliminary morphology analysis to predict tachycardia origin before delivering therapy. By anticipating whether the tachycardia is SVT or VT based on waveform characteristics, the system can preemptively select appropriate therapy (pacing for SVT, shock for VT), preventing unnecessary high-voltage shock delivery and its associated harms.
Solution Approach 2:
Morphology comparison is performed in advance of therapy delivery to classify the tachycardia type. This preliminary classification enables the system to prepare and deliver the most appropriate therapy (ATP for SVT, cardioversion for VT) without delay, avoiding unnecessary aggressive shock therapy and its harmful effects.
3Measurement precision
If morphology waveform matching is performed to distinguish SVT from VT, then classification accuracy is improved, but detection time and computational complexity increase
Solution Approach 1:
The morphology analysis compares only the critical portions of waveforms (P-wave, QRS complex, T-wave characteristics) rather than entire cardiac cycles. This partial action approach maintains high classification accuracy while significantly reducing computational time and complexity compared to full waveform analysis.
Solution Approach 2:
The system transforms waveform data into standardized morphological parameters (amplitude, duration, interval measurements) that enable rapid comparison. By changing the representation from raw continuous waveforms to discrete standardized parameters, the system achieves accurate SVT/VT differentiation with minimal computational overhead.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A medical device and associated method for classifying an unknown cardiac signal sensing a cardiac signal over a plurality of cardiac cycles using a plurality of electrodes coupled to a sensing module, determining a template of a known cardiac signal in response to the cardiac signal sensed over the plurality of cardiac cycles, sensing an unknown cardiac signal over an unknown cardiac cycle, determining a fourth order difference signal corresponding to the template and a fourth order difference signal of the unknown cardiac signal, determining a first morphology match metric between the template fourth order difference signal and the fourth order difference signal of the unknown cardiac signal, and classifying the unknown cardiac signal in responsed to the determined first morphology match score.