13 June 2022 in IEEE Journal of Biomedical and Health Informatics
Liver tumor segmentation plays an essential role in diagnosis and treatment of hepatocellular carcinoma or metastasis. However, accurate and automatic tumor segmentation remains a challenging task, owing to vague boundaries and large variations in shapes, sizes, and locations of liver tumors. In this paper, we propose a novel hybrid end-to-end network, called TD-Net, which incorporates Transformer and direction information into convolution network to segment liver tumor from CT images automatically. The proposed TD-Net is composed of a shared encoder, two decoding branches, four skip connections, and a direction guidance block. The shared encoder is utilized to extract multi-level feature information, and the two decoding branches are respectively designed to produce initial segmentation map and direction information. To preserve spatial information, four skip connections are used to concatenate each encoder layer and its corresponding decoder layer, and in the fourth skip connection a Transformer module is constructed to extract global context. Furthermore, a direction guidance block is well-designed to rectify feature maps to further improve segmentation accuracy. Extensive experiments conducted on public LiTS and 3DIRCADb datasets validate that the proposed TD-Net can effectively segment liver tumor from CT images in an end-to-end manner and its segmentation accuracy surpasses those of many existing methods.
6 May 2022 in Expert Systems with Applications
4 March 2024 in IEEE Transactions on Fuzzy Systems
Outdoor images taken in haze usually exhibit contrast reduction, color distortion, and detail loss. Removing the haze from a given image is a tough issue owing to its highly ill-posed property. To restore the haze-free image effectively, we develop an unsupervised dehazing method using patch-line and fuzzy clustering-line priors in this paper. The method obtains the dehazed image by inversely solving the atmospheric scattering model, which involves in estimating two key parameters, including atmospheric light and scene transmission. First, the orientation of atmospheric light is achieved by using a patch-line prior. Then, a quadtree subspace hierarchical searching scheme is designed to get the magnitude by calculating the differences between the mean and variance of each component for local regions. Besides, a fuzzy clustering-line prior combined with a guided filtering is proposed to estimate the scene transmission for each pixel. The proposed method can obtain the dehazed image directly without any training process and achieve much better performance than many existing ones with less space and time cost.
6 November 2023 in Optics & Laser Technology
1 September 2025 in IEEE Transactions on Image Processing
CNNs have demonstrated superior performance in medical image segmentation. To overcome the limitation of only using local receptive field, previous work has attempted to integrate Transformers into convolutional network components such as encoders, decoders, or skip connections. However, these methods can only establish long-distance dependencies for some specific patterns and usually neglect the loss of fine-grained details during downsampling in multi-scale feature extraction. To address the issues, we present a novel hybrid Transformer network called FocalTransNet. specifically, we construct a focal-enhanced (FE) Transformer module by introducing dense cross-connections into a CNN-Transformer dual-path structure and deploy the FE Transformer throughout the entire encoder. Different from existing hybrid networks that employ embedding or stacking strategies, the proposed model allows for a comprehensive extraction and deep fusion of both local and global features at different scales. Besides, we propose a symmetric patch merging (SPM) module for downsampling, which can retain the fine-grained details by stablishing a specific information compensation mechanism. We evaluated the proposed method on four different medical image segmentation benchmarks. The proposed method outperforms previous state-of-the-art convolutional networks, Transformers, and hybrid networks. The code for FocalTransNet is publicly available at https://github.com/nemanjajoe/FocalTransNet.