Cardiac Vectogram Noise Filtering for Implantable Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Active implantable medical devices are prone to detecting non-cardiac signals as noise, leading to inappropriate therapies such as defibrillation shocks, which can be painful and harmful, due to issues like electromagnetic interference and myopotentials, and existing methods for filtering noise are either invasive or ineffective.
Innovation Solution
The device analyzes endocardial electrogram signals from two distinct channels to construct a two-dimensional cardiac loop, determining the intrinsic morphology of the signals and classifying cardiac cycles as valid or noisy based on descriptor parameters like the average angle formed by consecutive velocity vectors, thereby distinguishing between cardiac and non-cardiac signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If common filtering methods (blanking period, sensitivity control) are used to reduce noise detection, then false detections may be reduced, but the device complexity increases and effectiveness is limited
Solution Approach 1:
The patent transforms the one-dimensional temporal signal analysis into two-dimensional spatial analysis by constructing cardiac loops in the plane formed by bipolar and unipolar signals. This dimensional transformation enables the device to distinguish cardiac from non-cardiac signals based on their geometric patterns rather than relying on complex temporal filtering algorithms, thereby improving detection accuracy without significantly increasing device complexity
Solution Approach 2:
The patent replaces complex electronic filtering mechanisms (blanking periods, sensitivity adjustments) with a geometric pattern recognition approach. By analyzing the spatial configuration of cardiac loops and their velocity vectors, the system substitutes mechanical filtering complexity with mathematical morphology analysis, achieving better noise rejection with simpler implementation
2Reliability
If external electrodes are added to detect and subtract noise signals, then noise filtering improves, but the mechanical and electrical structure of the device must be redesigned
Solution Approach 1:
The patent extracts the noise detection capability from the existing bipolar and unipolar signal channels without adding external electrodes. By analyzing the geometric patterns (cardiac loops) formed by the existing signal combinations, the system separates cardiac signals from non-cardiac interference purely through mathematical transformation, eliminating the need for additional hardware components
Solution Approach 2:
The patent makes the existing signal channels serve multiple functions: the bipolar and unipolar signals are used both for their traditional pacing and sensing functions, and simultaneously for constructing cardiac loops to identify and filter non-cardiac signals. This multi-functionality approach allows noise filtering without requiring dedicated external electrodes or additional hardware
3Reliability
If defibrillation shocks are delivered to treat detected arrhythmias, then life-threatening conditions are treated, but the patient experiences pain and potential side effects
Solution Approach 1:
The patent implements a feedback mechanism where the morphology analysis of cardiac loops continuously monitors signal characteristics. By analyzing the geometric patterns and velocity vectors of successive cardiac loops, the system provides feedback to distinguish true arrhythmias from noise, ensuring that defibrillation shocks are delivered only when genuinely indicated, thereby reducing inappropriate treatments and their associated harm
Data Source
AI summary
Apparatus and method for detecting and filtering noise artifacts by analysis of a cardiac vectogram is disclosed. An active medical device collects electrical activity signals of a patient's heart over a series of cardiac cycles. At least two distinct temporal components (Vbip, Vuni) are obtained from at least two endocardial electrogram (EGM) signals that are collected concurrently on different respective channels from the same heart cavity. The means for analyzing and filtering of the active medical device operates by: constructing a non-temporal 2D characteristic of a vectogram (VGM) of an analyzed cardiac cycle, using a variation of one of the two distinct temporal components (Vuni) as a function of the other (Vbip); conducting a morphological analysis of the non-temporal 2D characteristic to derive at least one intrinsic descriptor parameter, including the average angle of velocity vectors of consecutive data points of the 2D characteristic, and classifying the analyzed cardiac cycle as invalid noisy cycle or as valid not noisy cycle, depending on the analysis of the descriptor parameter.