QRS Detection in Compressed ECG Signals at High Compression Ratios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing QRS detection algorithms are not designed for compressively sensed ECG data, leading to poor performance, especially at higher compression ratios, and lack tailored methods for signal reconstruction in electrocardiography.
Innovation Solution
A method and system for detecting QRS complexes in ECG signals using compressively sensed measurements, involving signal reconstruction algorithms that construct an estimate of the ECG signal, compute the first-order difference, and process it to locate significant natural blocks indicating QRS complexes, allowing for effective detection even at high compression ratios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing QRS detection algorithms (Pan-Tompkins, Hamilton-Tompkins) are applied to compressively sensed ECG data, then the detection process can be performed, but the detection accuracy deteriorates significantly, especially at compression ratios exceeding 71%
Solution Approach 1:
The patent transforms the QRS detection problem from the time domain to the frequency domain by applying Fourier transform. This parameter transformation allows the detection algorithm to operate effectively on compressively sensed data by exploiting the spectral characteristics of QRS complexes, which remain distinguishable even after compression. The frequency domain representation preserves critical detection features while being robust to the effects of compressive sensing.
Solution Approach 2:
The patent replaces traditional time-domain signal processing mechanisms with frequency-domain analysis. Instead of using temporal filtering and derivative operations on reconstructed signals, the invention uses spectral analysis and frequency domain filtering to detect QRS complexes directly from compressed measurements, substituting mechanical signal reconstruction with a more robust frequency-based approach.
2Measurement precision
If existing QRS detection algorithms are used on uncompressed ECG data, then detection accuracy is maintained, but the data transmission and storage requirements increase
Solution Approach 1:
The patent extracts only the essential spectral features needed for QRS detection from the compressively sensed data, rather than reconstructing and transmitting the entire ECG signal. By working directly in the frequency domain with compressed measurements, the algorithm extracts detection-critical information while discarding redundant data, achieving accurate detection with minimal data volume.
Solution Approach 2:
The patent performs frequency domain transformation and detection feature extraction on the compressed data before any reconstruction attempt. This preliminary processing in the frequency domain prepares the data for efficient detection, allowing the system to achieve accurate QRS identification without the need for full signal reconstruction, thereby minimizing data requirements.
3Productivity
If ECG signals are reconstructed from compressively sensed data using existing methods, then signal recovery can be achieved, but the reconstruction quality deteriorates at higher compression ratios
Solution Approach 1:
The patent applies frequency domain transformation as a preliminary step before detection, transforming the compressively sensed data into the frequency domain where QRS complexes exhibit distinct spectral characteristics. This preliminary transformation enables the detection algorithm to identify QRS complexes based on their frequency signatures rather than relying on accurate time-domain signal reconstruction, thereby maintaining detection quality at high compression ratios.
Solution Approach 2:
The patent substitutes traditional time-domain signal reconstruction mechanisms with frequency-domain detection. Instead of attempting to recover the time-domain signal waveform through compression algorithms, the invention directly analyzes the spectral content of compressed measurements to detect QRS complexes, replacing the reconstruction mechanism with a more robust frequency-based detection approach.
Data Source
AI summary
Electrocardiogram (ECG) data is compressible at high compression ratios using suitable compressive sensing techniques. Methods of detecting QRS complexes in an ECG signal may comprise receiving compressively-sensed measurements of an ECG signal; constructing an estimate of the ECG signal from the received compressively-sensed measurements, and detecting QRS complexes in the estimate of the ECG signal. QRS complexes may be detected by computing the first-order difference of the estimate of the ECG signal and processing the first-order difference of the estimate of the ECG signal to locate one or more significant natural blocks, each indicating a QRS complex in the ECG signal. QRS complexes may also be detected by using a conventional QRS detection algorithm on the estimate of the ECG signal. Also disclosed are related systems for detecting QRS complexes and for compressively sensing ECG signals.