Breast mass segmentation method based on multi-modal feature fusion Vnet

A feature fusion and multi-modal technology, applied in the field of medical image processing, can solve the problems that small lesions cannot be accurately judged, places with too small pixel contrast cannot be well resolved, and the diagnostic accuracy is reduced, so as to avoid incompleteness Sex and active bias, reducing prior knowledge dependence, and improving accuracy

Pending Publication Date: 2022-05-27
NANTONG UNIVERSITY
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although breast magnetic resonance imaging is an important method for the examination and diagnosis of breast diseases, it is not easy for doctors to read MRI images. Heavy work, visual fatigue, which caused a decline in diagnostic accuracy
In the p

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
  • Breast mass segmentation method based on multi-modal feature fusion Vnet
  • Breast mass segmentation method based on multi-modal feature fusion Vnet
  • Breast mass segmentation method based on multi-modal feature fusion Vnet

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following detailed description is given in conjunction with the accompanying drawings and specific embodiments.

[0059] like figure 1 As shown, the present invention provides a breast mass segmentation method based on multimodal feature fusion Vnet, comprising the following steps:

[0060] S1: A dataset of breast magnetic resonance images and the results of breast tumor segmentation manually marked by doctors, mainly including four different modalities of breast dynamic scan magnetic resonance images, namely dynamic scan phase 1, dynamic scan phase 2, and dynamic scan 4. period, 6 periods of dynamic scanning;

[0061] S2: data preprocessing, dividing the data in the data set;

[0062] S3: Construct a network model based on multimodal feature fusion Vnet;

[0063] S4: Train the network model in step S3, adjust parameters, and obtain the predi...

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 multi-modal feature fusion Vnet-based breast mass segmentation method. The method comprises the following steps: S1, obtaining a breast magnetic resonance image and a data set of breast mass segmentation results manually marked by a doctor; s2, data preprocessing: dividing data in the data set; s3, constructing a network model based on multi-modal feature fusion Vnet; s4, training the network model in the step S3, and performing parameter adjustment to obtain a predicted breast lump segmentation result; and S5, comparing the predicted breast lump segmentation result obtained in the step S4 with the breast lump segmentation result manually marked by the doctor in the step S1 by using a set evaluation index and a loss function, and verifying the effectiveness of the segmentation method. The breast mass segmentation accuracy can be effectively improved, doctors are assisted in diagnosis and decision making, the burden of the doctors is relieved, and the method has high application value in breast mass auxiliary diagnosis, operation simulation and medical teaching.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a breast mass segmentation method based on multimodal feature fusion Vnet. Background technique [0002] Breast cancer is the most common malignancy in women and the leading cause of cancer death in women worldwide. Early screening and treatment are important means to improve the survival rate of breast cancer patients. Among them, magnetic resonance imaging technology has the characteristics of excellent soft tissue resolution and no radiation, and has unique advantages in breast examination. In recent years, more and more clinical and research work has been carried out on breast MRI in my country. Among high-risk patients with breast cancer, breast MRI has higher sensitivity. Although breast MRI is an important method for the examination and diagnosis of breast diseases, it is not easy for doctors to read MRI images. The doctor's judgment process is often time-consuming...

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
IPC IPC(8): G06T7/11G06K9/62G06N3/04G06N3/08G06V10/80G06V10/82
CPCG06T7/11G06N3/08G06T2207/10088G06T2207/20132G06T2207/30068G06N3/045G06F18/253
Inventor 高瞻陈蓉邵叶秦王杰华
Owner NANTONG UNIVERSITY
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