Rotor Pivot Point Mapping via Multi-Scale Signal Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying rotor pivot points in the heart, particularly for atrial fibrillation, are limited in accuracy due to noise and misleading phase and activation times, leading to suboptimal catheter ablation success rates in patients with persistent AF, as existing electro-anatomic mapping systems fail to predict rotor locations outside pulmonary veins.
Innovation Solution
The use of multi-scale frequency (MSF), kurtosis, empirical mode decomposition (EMD), and multi-scale entropy (MSE) calculations to generate datasets that graphically indicate pivot points of rotors, enabling more precise spatiotemporal mapping and guiding patient-specific ablation therapy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If known signal processing techniques (DF, CFAE, LAT) are used for rotor identification, then the mapping process is simple and familiar, but the accuracy of predicting rotor locations outside pulmonary veins deteriorates
Solution Approach 1:
The patent transforms electrogram signals from the time domain to the frequency domain using Fast Fourier Transform (FFT), and then applies Hilbert transform to extract instantaneous frequency and phase information. This parameter transformation enables accurate identification of rotor locations by analyzing frequency components and phase singularities, resolving the contradiction between measurement precision and method complexity.
Solution Approach 2:
The patent replaces traditional mechanical/electrical mapping approaches with computational signal processing methods. By using mathematical transformations (FFT, Hilbert transform) to analyze electrogram data, the system achieves high-precision rotor localization without requiring complex physical measurement devices, thus improving accuracy while managing computational complexity.
2Measurement precision
If clinical electrogram signals are used directly for mapping, then the data acquisition is straightforward, but the signals may not represent local activation and distort rotor location prediction
Solution Approach 1:
The patent introduces an intermediary computational processing stage between signal acquisition and rotor identification. The Fast Fourier Transform and Hilbert transform act as intermediaries that convert raw electrogram signals into frequency and phase domain representations, making the local activation patterns more distinguishable and reducing distortion in rotor location prediction.
Solution Approach 2:
The patent transitions from analyzing signals in the time domain to analyzing them in the frequency domain using FFT, and further to the instantaneous frequency and phase domain using Hilbert transform. This dimensional transformation reveals hidden patterns in the electrogram signals that are not apparent in the time domain, improving the accuracy of local activation representation while providing structured methods for signal interpretation.
3Area of stationary object
If virtual electrograms from non-contact methods are used, then the spatial coverage is expanded, but information distortion occurs in AF analysis
Solution Approach 1:
The patent employs phase singularities as feedback indicators to identify rotor locations. By analyzing the phase information extracted through Hilbert transform, the system detects points where phase wraps around, which correspond to rotor cores. This feedback mechanism enables accurate rotor identification even with expanded spatial coverage from non-contact mapping methods, counteracting information distortion.
Data Source
AI summary
Disclosed herein are techniques for graphically indicating aspects of rotors (such as pivot points of rotors) associated with atrial or ventricular fibrillation. Embodiments can include receiving, using a processor, an electrogram for each of a plurality of spatial locations in a heart, each electrogram comprising time series data including a plurality of electrical potential readings over time. Embodiments can also include generating, from the time series data, one or more of multi-scale frequency (MSF), kurtosis, empirical mode decomposition (EMD), and multi-scale entropy (MSE) datasets, each dataset including a plurality of respective values or levels corresponding to the plurality of spatial locations in the heart. Also, examples can include generating, from the one or more datasets, a map including a plurality of graphical indications of the values or levels at the plurality of spatial locations in the heart, wherein the map can include an image of the heart and graphical indications of locations of aspects of rotors in the heart (such as pivot point of rotors).