Related Experiment Video
Updated: Jun 26, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.7K
ConvMedSegNet: A multi-receptive field depthwise convolutional neural network for medical image segmentation
Yuxu Peng1, Xin Yi1, Dengyong Zhang1
1School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha, 410114, China.
Computers in Biology and Medicine
|May 17, 2024
Summary
This study introduces ConvMedSegNet, a novel deep learning model for precise medical image segmentation. It effectively captures multi-scale textures and fuses features to improve accuracy in medical imaging tasks.
Area of Science:
- Medical Imaging
- Computer Vision
- Deep Learning
Background:
- Accurate medical image segmentation is crucial for diagnosis and treatment planning.
- Existing methods often struggle with capturing multi-scale features and preserving global information.
Purpose of the Study:
- To develop a novel convolutional neural network, ConvMedSegNet, for highly precise medical image segmentation.
- To enhance feature representation by integrating multi-scale information and improving feature fusion.
Main Methods:
- Proposed ConvMedSegNet with a U-shaped architecture.
- Introduced a multi-receptive field depthwise convolution (MRDC) module to capture multi-scale textures.
- Implemented a guided fusion (GF) module for effective encoder-decoder feature fusion.
Main Results:
- ConvMedSegNet demonstrated superior performance compared to advanced methods on BUSI and ISIC2018 datasets.
- The MRDC module effectively captured varying texture information, enhancing global feature correlation.
- The GF module minimized critical data loss during multi-scale feature fusion.
Conclusions:
- ConvMedSegNet achieves state-of-the-art results in medical image segmentation.
- The proposed modules effectively address limitations in capturing multi-scale features and feature fusion.
- The model offers a promising solution for improving the precision of medical image analysis.
