Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Neural network training method and convolutional neural network for premature ventricular contraction heartbeat positioning

A convolutional neural network and training method technology, applied in the field of electrocardiogram processing, can solve problems such as low accuracy, complex diagnosis methods, and inability to obtain the onset time of premature ventricular contractions

Inactive Publication Date: 2019-07-23
SHANGHAI SID MEDICAL CO LTD
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The diagnostic accuracy of this type of method is not only related to the heartbeat diagnostic algorithm, but also affected by the R wave detection algorithm, which leads to many disadvantages such as complex diagnostic methods and low accuracy.
[0004] Chinese patent document CN108511055A discloses a system and method for identifying premature ventricular beats based on classifier fusion and diagnostic rules. Two classifiers, convolutional neural network and recursive neural network, are used to identify premature ventricular beats. However, this The application cannot avoid the impact of R wave detection on the recognition accuracy of premature ventricular contractions, and the technical solution of the application cannot obtain the start time of premature ventricular contractions

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Neural network training method and convolutional neural network for premature ventricular contraction heartbeat positioning
  • Neural network training method and convolutional neural network for premature ventricular contraction heartbeat positioning
  • Neural network training method and convolutional neural network for premature ventricular contraction heartbeat positioning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0028] This embodiment provides an improved convolutional neural network training method for ventricular premature heartbeat positioning, such as image 3 As shown, including the following steps:

[0029] S1: Acquire ECG signals known as premature ventricular beats and other heartbeats of uniform type. The ECG data can also be preprocessed. The ECG data is filtered by fir filters with upper and lower cutoff frequencies of 0.1 Hz and 100 Hz respectively. , If the ECG signal sampling frequency is not 500Hz, the nearest neighbor interpolation method will be used to resample the ECG signal to 500Hz;

[0030] S2: Use a sliding window with a step length of 0.015-0.025s and a length of 0.4-0.8s to intercept the ECG signal to form an interception segment. The length is preferably 0.6s. 0.4-0.8s corresponds to the length of a heartbeat cycle. Set the step size to 0.015 -0.025s can effectively avoid the influence of R wave, meet the clinical accuracy and at the same time help reduce the amou...

Embodiment 2

[0037] This embodiment provides an improved convolutional neural network for ventricular premature beat positioning, which is obtained by training the improved convolutional neural network training method for ventricular premature beat positioning of Embodiment 1.

[0038] The use method of the improved convolutional neural network for ventricular premature heartbeat positioning in embodiment 2 is:

[0039] The ECG signal of unknown heartbeat type is intercepted by a sliding window with a step length of 0.015-0.025s and a length of 0.4-0.8s to form an intercepted segment, and all the intercepted segments are introduced into the improved convolutional nerve for ventricular early heartbeat positioning In the network, obtain the output value of the improved convolutional neural network. If the output value is greater than the midpoint value of the continuous value (the midpoint value is 0.5 when the output value is [0,1]), it is considered that the ECG signal With ventricular prematur...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a premature ventricular contraction heartbeat positioning method and device based on an improved convolutional neural network. The method comprises the following steps: acquiring an ECG signal, intercepting the ECG signal by using a sliding window with a step size of 0.015-0.025s and a length of 0.4-0.8s to form an intercepted section, and training a convolutional neural network with the intercepted section to form a ventricular premature beat heartbeat judgment model. The 0.4-0.8s just corresponds to the length of one heartbeat period, and the step size of 0.015-0.025s is set to effectively avoid the influence of R waves so as to meet the judgment precision on ventricular premature beat heartbeat, and to facilitate the reduction of calculation amount.

Description

Technical field [0001] This application belongs to the technical field of electrocardiogram processing, and in particular relates to a method and device for locating premature ventricular beats based on an improved convolutional neural network. Background technique [0002] Before the sinoatrial node impulse reaches the ventricle, an electrical impulse is sent out in advance from any part of the ventricle or the ectopic rhythm point of the ventricular septum to cause the depolarization of the ventricle, which is called ventricular premature contraction, or premature ventricular for short. Although the incidental room seen in normal healthy people is not clinically significant as early as, but in the case of the examiner suffering from organic heart disease, it must be combined with clinical symptoms and medical history, analyzed according to different situations and given necessary treatment. Especially for patients with Holter monitoring, it is necessary to select the room earl...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): A61B5/0402A61B5/0456A61B5/352
CPCA61B5/7267A61B5/352A61B5/318
Inventor 朱俊江严天宏杨旭堃何雨辰李滨邓欣
Owner SHANGHAI SID MEDICAL CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com