Wearable ECG Pain Classifier Using SVM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems fail to accurately differentiate between cardiac and non-cardiac pain without surrogate signals and are limited to clinical settings, making it difficult to determine the origin of chest pain effectively.
Innovation Solution
A device and method that uses electrocardiogram (ECG) and heart rate features, processed by a support vector machine classifier, to classify pain as cardiac or non-cardiac, without requiring exercise data, and can be used in any setting through a wearable device and mobile interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current systems use surrogate signals (blood pressure, heartrate, temperature, respiratory rate) to determine exercise data, then they can characterize events as exercise-related or non-exercise related, but they require additional sensors and complex processing to accurately differentiate cardiac from non-cardiac pain
Solution Approach 1:
The patent extracts and focuses specifically on ECG features (ST segment, PQ segment, and time intervals) as the critical diagnostic elements for pain differentiation, eliminating the need for multiple surrogate signals. By isolating the most discriminative ECG parameters, the system achieves accurate cardiac vs. non-cardiac pain differentiation without requiring blood pressure sensors, temperature sensors, or respiratory rate monitors, thus reducing device complexity while maintaining measurement precision.
Solution Approach 2:
The system performs preliminary classification of chest pain using ECG features before determining the need for additional exercise testing. By pre-screening with ECG analysis, the system can identify cases that are clearly cardiac or non-cardiac in origin, avoiding unnecessary complex testing protocols and reducing overall system complexity for routine cases.
2Adaptability or versatility
If current systems are designed for clinical settings only, then they can provide controlled diagnostic environments, but they are not available outside of clinical settings and cannot provide patient-specific protocols
Solution Approach 1:
The patent designs a universal ECG-based pain differentiation system that functions reliably across multiple settings (clinical and non-clinical). The core ECG analysis algorithm is setting-agnostic, using standardized ECG features that can be acquired in any environment. The system adapts to different contexts by adjusting the complexity of exercise protocols while maintaining the same fundamental diagnostic criteria, thus achieving both versatility and reliability.
Solution Approach 2:
The system dynamically adjusts exercise protocol intensity and duration based on initial ECG findings and patient response. In clinical settings, more comprehensive protocols may be employed, while in non-clinical settings, simplified protocols are used. The ECG-based classification adapts in real-time to patient conditions, maintaining diagnostic reliability across varying environments and patient populations.
3Measurement precision
If the system uses patient-specific protocols with varying levels of difficulty, then it can provide more accurate diagnosis, but it requires complex instruction generation and adaptive control
Solution Approach 1:
The system generates exercise instructions with varying difficulty levels based on preliminary ECG assessment and patient characteristics. By pre-defining standardized protocol templates with different intensity levels, the system avoids complex real-time protocol design while still providing personalized testing. The ECG-based classification guides which pre-defined protocol level to use, simplifying operation while maintaining diagnostic accuracy.
4Measurement precision
If the system processes multiple ECG features (KLT components, time intervals) and activity level variations, then it can achieve accurate cardiac pain classification, but it requires sophisticated signal processing and computational resources
Solution Approach 1:
The patent segments the ECG signal analysis into distinct, manageable feature extraction modules: ST segment analysis using Karhunen-Loeve transform, PQ segment analysis, and time interval measurements (R-P, R-S, S-J). Each segment is processed independently with specialized algorithms, then results are integrated for final classification. This segmentation reduces computational complexity by breaking down the complex signal processing task into smaller, more efficient sub-tasks that can be executed with fewer computational resources.
Data Source
AI summary
A device, system, and method is provided for detecting pain in a cardiac-related region of the body and determining whether that pain is cardiac or non-cardiac. The device, system, and method may include calculating or determining a first feature based on a variation in activity level and a variation in the detected heartrate measurement and a second feature based on a variation in the detected ECG features and a first feature and then subjecting at least the first feature and the second feature to a cardiac pain classifier to determine a cardiac classification.