2D Vectogram Analysis for VT/SVT Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current algorithms for discriminating between Ventricular Tachycardia (VT) and Supra-Ventricular Tachycardia (SVT) in active implantable medical devices are prone to false diagnoses, leading to inappropriate therapy delivery, due to limitations in analyzing endocardial electrogram signals and high computational demands.
Innovation Solution
A bi-dimensional analysis method using EGM signals from two channels, represented as a 'vectogram' in a 2D space, compares current tachycardia beats to a reference Sinus Rhythm beat, employing geometrical descriptors like unit tangent vectors and curvature to differentiate between VT and SVT, while also considering eigenvalues and correlation coefficients for improved discrimination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional algorithms analyze endocardial electrogram signals to discriminate between VT and SVT, then the discrimination function is provided, but false diagnoses occur leading to inappropriate therapy delivery
Solution Approach 1:
The patent transforms the traditional one-dimensional temporal analysis of EGM signals into a two-dimensional phase plane analysis by plotting dV/dt versus V. This dimensional transformation creates a new analytical space where VT and SVT exhibit distinct geometric patterns, enabling more reliable discrimination. The phase plane representation reveals characteristic loop morphologies that are not apparent in conventional unipolar EGM waveforms alone.
Solution Approach 2:
The patent employs geometrical descriptors to characterize the shape, orientation, and morphology of the phase plane loops, analogous to using color as a distinguishing feature. By analyzing loop area, eccentricity, orientation angles, and other geometric properties, the system identifies distinctive patterns that differentiate VT from SVT, similar to how color changes can indicate different states or conditions.
2Measurement precision
If high computational algorithms are used to improve discrimination accuracy, then measurement precision improves, but device complexity and computational demands increase
Solution Approach 1:
The patent segments the complex discrimination problem into distinct geometric feature extractions from the phase plane loops. By calculating separate descriptors such as loop area, eccentricity, orientation, and morphology parameters, the system breaks down the analysis into manageable components that can be processed independently and then integrated for final classification.
Solution Approach 2:
The patent transforms the raw EGM signal parameters into derived geometric parameters of the phase plane loops. By changing from temporal domain analysis to spatial domain analysis of the phase plane, the system creates new parameters (loop area, eccentricity, orientation angles) that provide discriminative power while being computationally efficient to calculate and interpret.
3Reliability
If defibrillation shocks are delivered for all detected tachycardias, then therapy coverage is maximized, but harmful effects increase due to inappropriate shocks in SVT cases
Solution Approach 1:
The patent applies preliminary discrimination analysis using phase plane geometric features before committing to defibrillation therapy. By evaluating the characteristic loop patterns in advance, the system identifies SVT cases and prevents inappropriate defibrillation shocks, thereby avoiding the harmful effects of unnecessary high-energy deliveries while ensuring appropriate therapy for true VT cases.
Data Source
AI summary
An active medical device able to discriminate between tachycardias of ventricular origin and of supra-ventricular origin. Two distinct temporal components (UnipV, BipV) are obtained corresponding to two EGM signals of ventricular electrograms. The diagnosis operates in at least two-dimensional space to determine, from the variations of one temporal component as a function of the other temporal component, a 2D characteristic representative of a heart beat and, this, for a reference beat collected in Sinus Rhythm (SR) in the absence of tachycardia episodes, and for a heart beat in Tachycardia. The discrimination of the tachycardia type, VT or SVT, is then realized by a classifier operating a comparison of the two current and reference 2D characteristics.