Intracardiac ECG Signal Annotation via Gaussian Waveform Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
ECG signals from body organs are noisy due to various factors such as movement, electrode contact changes, and interference, making it difficult to accurately measure annotation times.
Innovation Solution
Fitting intracardiac signals to predetermined oscillating waveforms derived from Gaussian functions, with asymmetry factors, to reduce noise and variability in annotation times, allowing for more precise measurement of signal characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If ECG signals are measured during medical procedures, then electrical signals from body organs can be obtained, but noise is introduced due to movement, electrode contact changes, and interference
Solution Approach 1:
The patent extracts the essential signal characteristics by fitting the noisy ECG signal to a predetermined waveform model, separating the useful signal information from the noise through mathematical modeling and parameter extraction
Solution Approach 2:
The patent introduces a predetermined waveform model as an intermediary between the raw noisy signal and the final annotation, using the model's mathematical structure to filter noise while preserving signal characteristics
2Loss of time
If annotation times are measured directly from noisy ECG signals, then time measurements can be obtained, but measurement precision deteriorates due to signal variability
Solution Approach 1:
The patent performs preliminary waveform fitting and parameter optimization before extracting annotation times, preparing a cleaned and modeled version of the signal that yields more precise time measurements
Solution Approach 2:
The patent replaces direct measurement from raw signals with a mathematical modeling approach, substituting the mechanical/electrical measurement process with computational waveform analysis and parameter extraction
3Object-affected harmful factors
If noise reduction is applied to ECG signals, then signal clarity is improved, but measurement precision of annotation times may be affected
Solution Approach 1:
The patent uses iterative optimization to adjust waveform parameters, comparing the modeled signal against the actual signal and refining the fit to achieve both noise reduction and accurate annotation time extraction
Solution Approach 2:
The patent changes the parameters of the predetermined waveform model during fitting to match the characteristics of the actual ECG signal, optimizing the model to preserve true signal features while filtering noise
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for analysin ECG signals is disclosed. The method includes sensing a time-varying intracardiac potential signal and finding a fit of the time-varying intracardiac potential signal to a predefined oscillating waveform (step 206). The predefined oscillating waveform comprises a first differential of a Gaussian function, which is skewed by an asymmetry factor. An annotation time of the signal may then be derived from the fitted signal, rather than from the raw signal (step 212).