ECG Signal Peak Detection via Fractional Fourier Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing algorithms for detecting cardiovascular diseases from ECG signals are not accurate enough, are complex, and require intense computing capabilities, making them expensive and difficult to use effectively.
Innovation Solution
A system that uses a combination of two-event related moving averages and fractional-Fourier-transform (FrFT) to denoise ECG signals, followed by a machine learning classifier to identify cardiovascular diseases, which includes a wavelet transform for noise removal and a logic circuit for peak detection and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ECG algorithms (EMD-Hilbert transform, rapid-ramp) are used for peak detection, then peak detection can be performed, but the algorithms are complicated and require intense computing capabilities
Solution Approach 1:
The patent applies parameter changes by transforming the ECG signal into the fractional Fourier transform domain with variable order parameter alpha, allowing adaptive optimization of peak detection performance. The fractional Fourier transform order is adjusted to balance between noise suppression and peak detection accuracy, resolving the contradiction between measurement precision and computational complexity.
Solution Approach 2:
The patent replaces complex mechanical signal processing algorithms (EMD-Hilbert transform) with a mathematical transformation approach using fractional Fourier transform combined with thresholding. This substitution simplifies the computational process while maintaining peak detection accuracy, reducing the intensive computing capabilities required.
2Reliability
If complex algorithms with multiple processing blocks are used for ECG signal processing, then noise removal and peak detection can be performed, but computing costs increase and ease of use decreases
Solution Approach 1:
The patent merges multiple separate processing blocks (noise removal, peak detection, classification) into a unified fractional Fourier transform-based framework. By combining these functions into a single integrated approach with adjustable parameters, the system maintains processing reliability while significantly improving ease of use through simplified operation.
Solution Approach 2:
The fractional Fourier transform framework serves multiple functions simultaneously: it acts as a noise filtering mechanism, a peak detection tool, and a feature extraction method for classification. This multi-functionality reduces the need for separate complex algorithms, thereby reducing computing costs and improving ease of use while maintaining reliability.
3Productivity
If traditional ECG processing methods are used, then basic diagnosis can be performed, but diagnostic accuracy is insufficient for reliable cardiovascular disease detection
Solution Approach 1:
The patent implements feedback mechanisms through iterative optimization of the fractional Fourier transform order parameter and adaptive thresholding. The system continuously adjusts processing parameters based on signal characteristics to maintain high diagnostic accuracy while operating at high speed, resolving the contradiction between productivity and measurement precision.
Solution Approach 2:
The patent transitions from traditional time-domain ECG analysis to the fractional Fourier transform domain, adding a new dimensional perspective to signal processing. This dimensional change enables simultaneous achievement of high diagnostic accuracy and fast processing by capturing signal features that are not visible in the time domain alone.
Data Source
AI summary
A method for automatically and independently associating a cardiovascular disease with an electrocardiogram, ECG, signal includes receiving the ECG signal; denoising the ECG signal using a wavelet transform; determining peaks of the denoised ECG signal by applying a combination of (1) two event-related moving averages and (2) fractional-Fourier-transform (FrFT) to the denoised ECG signal; and passing the peaks through a classifier for identifying the cardiovascular disease that corresponds to the ECG signal.