Multi-path connected network for medical image segmentation |
| |
| Affiliation: | 1. School of Architecture, South China University of Technology, Guangzhou 510641, China;2. Foreign Language Teaching Department, Guang Zhou Vocational School of Finance and Economics, Guang Zhou 510080, China;3. School of Financial Mathematics and Statistics, GuangDong University of Finance, Guangzhou 510521, China;1. Department of Computer Science and Technology, Jilin University, Changchun 130012, China;2. Key Laboratory of Symbolic Computation and Knowledge Engineering for Ministry of Education, Jilin University, Changchun 130012, China;3. Editorial Department of Journal (Engineering and Technology Edition), Jilin University, Jilin, Changchun 130012, China;1. Department of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China;2. College of information, Liaoning University, Liaoning 110036, China;1. Academy for Engineering and Technology, Fudan University, Shanghai 200433, China;2. Shanghai Engineering Research Center of AI & Robotics, China;3. Engineering Research Center of AI & Robotics, Ministry of Education, China;4. School of Information Science and Technology, Fudan University, Shanghai 200433, China;5. Department of Industrial and Manufacturing Systems Engineering, University of Michigan-Dearborn, Dearborn, MI 48128, USA;1. Chang’an University National Engineering Laboratory for Highway Maintenance Equipment, Xi’an 710100, China;2. Yan’an University College of Mathematics and Computer Science, Yan’an 716000, China |
| |
| Abstract: | In recent years, deep learning has been successfully applied to medical image segmentation. However, as the network extends deeper, the consecutive downsampling operations will lead to more loss of spatial information. In addition, the limited data and diverse targets increase the difficulty for medical image segmentation. To address these issues, we propose a multi-path connected network (MCNet) for medical segmentation problems. It integrates multiple paths generated by pyramid pooling into the encoding phase to preserve semantic information and spatial details. We utilize multi-scale feature extractor block (MFE block) in the encoder to obtain large and multi-scale receptive fields. We evaluated MCNet on three medical datasets with different image modalities. The experimental results show that our method achieves better performance than the state-of-the-art approaches. Our model has strong feature learning ability and is robust to capture different scale targets. It can achieve satisfactory results while using only 0.98 million (M) parameters. |
| |
| Keywords: | Medical image segmentation Multi-path connections Convolutional neural networks Encoder-decoder structure |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
