Deep learning based automatic coronary artery disease detection method, system and equipment

A coronary artery disease and deep learning technology, applied in neural learning methods, image data processing, image enhancement and other directions, can solve the problems of neglecting spatial information, low pixel detection accuracy, and large amount of calculation, so as to shorten the diagnosis and treatment process, improve the The effect of lesion detection rate

Active Publication Date: 2018-07-13
北京红云视界技术有限公司 +1
View PDF6 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Using manual features is time-consuming and labor-intensive, and different features are used for different lesions
However, using a deep network based on the classification task performed pixel by pixel, this method has a considerable amount of calculation, resulting in the inability to detect lesions in medical images in real time, and the accuracy of pixel detection is low because it ignores the spatial information in the image

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
  • Deep learning based automatic coronary artery disease detection method, system and equipment
  • Deep learning based automatic coronary artery disease detection method, system and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] Embodiment 1 of the present invention provides an automatic detection method for coronary artery lesions based on deep learning. The method includes the following steps:

[0055] Step 1. Training steps, including:

[0056] S101. The text information processing module extracts the basic characteristics of the patient from the coronary artery lesion medical records stored in the comprehensive medical database, and uses the C4.5 decision tree algorithm to train a classification decision tree based on the extracted basic characteristics, and the classification decision tree outputs the lesion type information To the Dicom video processing module;

[0057] S102, the Dicom video processing module uses SSN (temporal-segment-networks) to filter key frames containing lesions from the cardiac angiography Dicom video corresponding to the lesion type information stored in the medical comprehensive database through the received lesion type information , and output keyframes to the...

Embodiment 2

[0094] Another aspect of the present invention also provides a deep learning-based automatic detection system 200 for coronary artery lesions, including: a training unit and a testing unit;

[0095] Training units, including:

[0096] The text information processing module is used to extract the basic characteristics of the patient from the coronary artery lesion medical records stored in the medical comprehensive database, and use the C4.5 decision tree algorithm to train a classification decision tree based on the extracted basic characteristics, and the classification decision tree combines the lesion type information Output to the Dicom video processing module;

[0097] The Dicom video processing module is used to filter key frames containing lesions from the cardiac angiography Dicom video corresponding to the lesion type information stored in the medical comprehensive database using the SSN through the received lesion type information, and output the key frames into the...

Embodiment 3

[0110] Such as figure 2 As shown, another aspect of the present invention also provides a deep learning-based automatic detection device 300 for coronary artery lesions, including the system 200 described in the second embodiment.

[0111] The present invention provides an automatic detection device for coronary artery lesions based on deep learning, which applies the object detection technology based on deep learning to the detection of coronary artery lesions. Applying machine learning-based text processing techniques to coronary artery lesion detection. Integrating text processing and image processing technology, the information of multiple modalities is fused for coronary artery lesion detection. Fully automate the process of coronary artery lesion detection, without manual participation in the detection process. The technical solution of the invention can detect the lesion in the coronary arteries of the heart in real time, and provide reference and help to doctors. C...

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 provides a deep learning based automatic coronary artery disease detection method, system and equipment. A deep learning based object detection technique is applied to coronary artery lesion detection through a training step and a testing step; a machine learning based text processing technique is applied to coronary artery lesion detection; by technical fusion of text processing andimage processing, information of multiple modes is fused and applied to coronary artery lesion detection; a coronary artery lesion detection process is automated completely without artificial participation. By adoption of the technical scheme, technical problems of low accuracy in pixel detection and failure in real-time detection of lesions in medical images are solved, and heart coronary arterylesions can be detected out in real time to provide references and assistance for doctors. Compared with other systems, the system has advantages that the lesion detection rate is increased remarkably, and a diagnosis and treatment process is shortened.

Description

technical field [0001] The present invention relates to the technical field of digital image target detection, in particular to a method, system and equipment for automatic detection of coronary artery lesions based on deep learning. Background technique [0002] Coronary artery disease is one of the diseases with the highest mortality rate in the world. Digital Silhouette Angiography (DSA) and (CT) are currently the main methods that can diagnose the degree of cardiac arterial disease. Generally, doctors send a catheter into the coronary sinus of the heart, and then release a contrast agent through the catheter to visualize the coronary arteries. After that, the doctor observes according to different body positions and finally confirms the location of the lesion. Due to the delay in the flow of contrast medium and other reasons, it is sometimes difficult to accurately judge the lesion of a blood vessel. Detecting regions of interest or lesions in medical images is a key s...

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): G06T7/00G06N3/08G06N3/04
CPCG06N3/084G06T7/0012G06T2207/30101G06T2207/20221G06T2207/20081G06T2207/10016G06N3/045
Inventor 徐波杜天明周文辉
Owner 北京红云视界技术有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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