The key highlights and insights of the content are:
The authors propose a novel parallel multi-resolution encoder-decoder network, called PMR-Net, for medical image segmentation.
The parallel multi-resolution encoder can extract and fuse multi-scale fine-grained local features in parallel for input images with different resolutions. The multi-resolution context encoder fuses the global context semantic features of different receptive fields to maintain the integrity of global information.
The parallel multi-resolution decoder can continuously supplement the global context features of low-resolution branches to the feature maps of high-resolution branches, effectively solving the problem of global context feature loss caused by upsampling.
Extensive experiments on five public datasets demonstrate that PMR-Net outperforms state-of-the-art methods in segmentation accuracy.
PMR-Net is a flexible network framework that can be adjusted by changing the number of network layers and parallel encoder-decoder branches to meet the requirements of different medical imaging scenarios.
다른 언어로
소스 콘텐츠 기반
arxiv.org
더 깊은 질문