Coronary Heart Disease Probability from Reconstructed ECG Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing coronary heart disease are invasive, expensive, and inconvenient, necessitating a more accessible and efficient way to estimate the probability of the disease.
Innovation Solution
A method and apparatus that utilize ECG signals from multiple leads, reconstructing them using trained models to identify abnormal signals, and determine coronary heart disease probability based on these signals, incorporating features like differences and ratios between original and reconstructed signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If coronary angiography is used for detection, then diagnostic accuracy is improved, but invasiveness and operational complexity increase
Solution Approach 1:
The patent uses machine learning models to create a virtual copy of the coronary angiography diagnostic process. Instead of performing actual invasive angiography, the system processes ECG signals through trained models (including convolutional neural networks and attention mechanisms) to generate diagnostic results that replicate the functionality of traditional angiography, thereby eliminating operational complexity while maintaining diagnostic capability
Solution Approach 2:
The patent replaces the mechanical invasive procedure of coronary angiography with an information processing system. The mechanical catheter-based imaging is substituted by computational models that analyze ECG signals, transforming a physical invasive process into a non-invasive digital diagnostic approach while preserving essential diagnostic functions
2Measurement precision
If invasive detection methods like coronary angiography are used, then diagnostic accuracy is improved, but cost increases
Solution Approach 1:
The patent employs computationally efficient machine learning models that can be deployed on standard computing hardware, replacing expensive medical imaging equipment and consumables. The system uses lightweight neural network architectures that require minimal computational resources, making the diagnostic process cost-effective while maintaining accuracy through sophisticated signal analysis
3Ease of operation
If traditional ECG analysis methods are used, then convenience is improved, but diagnostic capability worsens
Solution Approach 1:
The patent transforms the diagnostic approach by changing the parameters analyzed in ECG signals. Instead of traditional visual inspection, the system extracts multiple features including time-domain, frequency-domain, and nonlinear dynamics parameters, then processes them through machine learning models. This parameter transformation enables standard ECG equipment to provide enhanced diagnostic capability while maintaining operational convenience
Solution Approach 2:
The patent performs preliminary feature extraction and signal processing on ECG data before final diagnosis. The system pre-processes ECG signals by detecting R-peaks, extracting waveforms, computing derivatives, and generating multiple feature sets in advance. This preliminary action prepares the data for efficient machine learning analysis, enhancing diagnostic capability without adding operational complexity for the end user
Data Source
Figure 1A~1I
Figure 2A~2I
Figure 3
AI summary
A method and an apparatus for determining a coronary heart disease probability are provided. The method includes: obtaining ECG signals of one or more leads of a user; determining, based on the ECG signals of the one or more leads, reconstructed signals respectively corresponding to the ECG signals of the one or more leads; and determining a coronary heart disease probability of the user based on the ECG signals of the one or more leads and the reconstructed signals respectively corresponding to the ECG signals of the one or more leads. In this way, a coronary heart disease probability of a user is estimated.