ECG Signal Classification via Symbolic Dynamics and Shannon Entropy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for processing electrocardiogram (ECG) signals are complex and inefficient, requiring large amounts of sample data and feature construction to train classifiers, which complicates accurate classification for clinical monitoring and telemedicine applications.
Innovation Solution
A method that converts the interval sequence of ECG signals into a symbol value sequence using symbolic dynamics and calculates the Shannon entropy to classify the signals, simplifying the process and improving calculation speed while discarding irrelevant noise, thereby measuring uncertainty more accurately without the need for training a classifier.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a classifier is trained using large amounts of sample data and feature construction, then classification accuracy may be improved, but the process becomes complicated and less efficient
Solution Approach 1:
The patent extracts only the essential temporal interval information from ECG signals, converting complex waveforms into simple interval sequences. This extraction approach discards irrelevant noise and detailed waveform characteristics while retaining the core temporal patterns needed for classification, thereby simplifying the processing pipeline without sacrificing diagnostic accuracy
Solution Approach 2:
The patent transforms the ECG signal representation from continuous amplitude-time waveforms into discrete interval sequences through parameter transformation. By changing the representation parameters from amplitude values to time intervals between waves, the method simplifies the data structure and enables more efficient processing while maintaining the essential temporal information needed for accurate classification
2Measurement precision
If traditional classification methods are used with feature construction and classifier training, then comprehensive analysis is achieved, but calculation speed decreases
Solution Approach 1:
The method extracts only the critical temporal interval features from ECG signals, converting complex waveforms into simple interval sequences. This extraction discards irrelevant noise and detailed waveform characteristics while retaining the core temporal patterns needed for classification, thereby simplifying the processing pipeline and enabling faster calculation
Solution Approach 2:
The patent uses symbolic dynamics to convert interval sequences into symbol value sequences, creating a simplified representation that copies the essential temporal patterns in a discrete, computationally efficient format. This symbolic copying enables rapid processing while preserving the diagnostic information needed for accurate classification
3Measurement precision
If large amounts of sample data are prepared for classifier training, then classification performance is improved, but the process becomes less efficient
Solution Approach 1:
The patent extracts only the essential temporal interval information from ECG signals, converting complex waveforms into simple interval sequences. This extraction approach discards irrelevant noise and detailed waveform characteristics while retaining the core temporal patterns needed for classification, thereby simplifying the processing pipeline without sacrificing diagnostic accuracy
Solution Approach 2:
The patent transforms the ECG signal representation from continuous amplitude-time waveforms into discrete interval sequences through parameter transformation. By changing the representation parameters from amplitude values to time intervals between waves, the method simplifies the data structure and enables more efficient processing while maintaining the essential temporal information needed for accurate classification
Data Source
AI summary
The present disclosure provides a method and apparatus for processing an electrocardiogram signal and an electronic device. The method includes: obtaining an electrocardiogram signal containing a plurality of selected waves; determining an interval sequence for the plurality of selected waves; converting the interval sequence into a symbol value sequence, based on symbolic dynamics; determining a Shannon entropy of the symbol value sequence; and classifying the electrocardiogram signal, based on a value of the Shannon entropy.