Parametric ECG Wave Analysis Using Cosine Phase Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing cardiac activity signals, such as Fourier decomposition and wavelet or Gaussian modeling, require a large number of parameters and lack physical meaning, making it difficult to characterize ECG signals effectively.
Innovation Solution
A method that analyzes cardiac activity signals by decomposing them into a sum of elementary waves expressed as x(t) = x0 + x1 cos(Φ(t)), where Φ(t) is the phase function, allowing for characterization with a small number of parameters that carry physical meaning and represent the shape of the signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Fourier decomposition is used to analyze ECG signals, then the frequency components of the signal can be described, but a large number of coefficients are required and they lack physical meaning
Solution Approach 1:
The patent transforms the ECG signal representation from Fourier coefficients to a parametric model with physically meaningful parameters (amplitude, duration, morphology parameters). This changes the parameter space from abstract mathematical coefficients to clinically interpretable quantities that directly describe wave characteristics.
Solution Approach 2:
The patent extracts and models only the essential characteristics of ECG waves (P, QRS, T waves) using a simplified parametric equation, separating the critical information from the redundant details that Fourier decomposition captures but cannot interpret meaningfully.
2Reliability
If wavelet or Gaussian modeling is used to decompose ECG signals, then the signal can be modeled, but a very large number of parameters are required for sufficient quality
Solution Approach 1:
The patent adopts a cosine-based parametric model with a small set of physically meaningful parameters instead of wavelet or Gaussian models. This parameter transformation achieves reliable ECG wave characterization while dramatically reducing the number of parameters needed for sufficient modeling quality.
3Loss of information
If Fourier decomposition is used, then the distribution of frequency components can be described, but no information is provided on the instants of appearance and wave shapes
Solution Approach 1:
The patent changes from frequency-domain Fourier coefficients to time-domain parametric models that explicitly represent wave instants, shapes, and morphologies. This parameter transformation preserves temporal and morphological information while maintaining mathematical simplicity.
Data Source
Figure 1~3
AI summary
The invention relates to a method for analysing the cardiac activity of a patient, comprising the following steps: acquisition (20) of at least one electric cardiac signal including at least one elementary signal corresponding to a heart beat; extraction (29), from the elementary signal, of at least one elementary wave having a general form that can be expressed as x(t) = x 0 + x 1 cos(F(t)), in which F(t) is the phase of the elementary wave; and analysis (30) of the elementary wave, comprising steps consisting in determining an expression of a phase equation, formula (I), of the elementary wave and determining an expression of phase F(t) of the elementary wave as a function of parameters measuring the anharmonicity of the elementary wave and the morphology thereof from functions pcosn and psinn defined by formula (II).