Method for detecting myocardial infarction based on wavelet packet features of short-time HRV signal
A detection method and wavelet packet technology, applied in the field of pattern recognition, can solve problems such as being unable to be widely used, and achieve the effects of easy operation, high efficiency, and clear principles
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0072] In this example, a total of 90 samples were used, including 45 healthy samples and 45 abnormal samples. A conventional 12-lead electrocardiogram was used, with a total of 1080 samples of 12*90. The seven extracted feature values were fused according to a ratio of 2:1. Divide the samples into 720 training samples and 360 test samples for training and testing, respectively use KNN for testing, the test accuracy rate is 74.4%; use ELM for testing, the test accuracy rate is 83.3%; use random forest for testing, the test accuracy The rate is 81.1%; using ensemble learning to test, the test accuracy rate is 75%. Among them, ELM and random forest have higher classification accuracy. Overall, the whole prediction method has a clear principle and high efficiency, and can identify the category of the sample more accurately in a shorter time. Figure 5 It is the classification recognition result of wavelet packet feature data in the embodiment of the present invention.
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap