Transformers have been widely used in many computer vision challenges and have shown the capability of producing better results than convolutional neural networks (CNNs). Taking advantage of capturing long-range contextual information and learning more complex relations in the image data, Transformers have been used and applied to histopathological image processing tasks. In this survey, we make an effort to present a thorough analysis of the uses of Transformers in histopathological image analysis, covering several topics, from the newly built Transformer models to unresolved challenges. To be more precise, we first begin by outlining the fundamental principles of the attention mechanism included in Transformer models and other key frameworks. Second, we analyze Transformer-based applications in the histopathological imaging domain and provide a thorough evaluation of more than 100 research publications across different downstream tasks to cover the most recent innovations, including survival analysis and prediction, segmentation, classification, detection, and representation. Within this survey work, we also compare the performance of CNN-based techniques to Transformers based on recently published papers, highlight major challenges, and provide interesting future research directions. Despite the outstanding performance of the Transformer-based architectures in a number of papers reviewed in this survey, we anticipate that further improvements and exploration of Transformers in the histopathological imaging domain are still required in the future. We hope that this survey paper will give readers in this field of study a thorough understanding of Transformer-based techniques in histopathological image analysis, and an up-to-date paper list summary will be provided athttps://github.com/S-domain/Survey-Paper.

Medical image segmentation is a critical task for the early detection and diagnosis of various conditions, such as skin cancer, polyps, thyroid nodules, and pancreatic tumors. Recently, deep learning architectures have achieved significant success in this field. However, they face a critical trade-off between local feature extraction and global context modeling. To address this limitation, we present DCM-Net, a dual-encoder architecture that integrates pretrained CNN layers with Visual State Space (VSS) blocks through a Cross-Branch Feature Fusion Module (CBFFM). A Decoder Feature Enhancement Module (DFEM) combines depth-wise separable convolutions with MLP-based semantic rectification to extract enhanced decoded features and improve the segmentation performance. Additionally, we present a new 2D pancreas and pancreatic tumor dataset (CCH-PCT-CT) collected from Chongqing University Cancer Hospital, comprising 3,547 annotated CT slices, which is used to validate the proposed model. The proposed DCM-Net architecture achieves competitive performance across all datasets investigated in this study. We develop a novel DCM-Net architecture that generates robust features for tumor and organ segmentation in medical images. DCM-Net significantly outperforms all baseline models in segmentation tasks, with higher Dice Similarity Coefficient (DSC) and mean Intersection over Union (mIoU) scores. Its robustness confirms strong potential for clinical use.

Accurate segmentation of tumor or organ regions in ultrasound images is essential for cancer diagnosis and treatment planning in biomedical engineering. However, the diversity and noisy nature of ultrasound images pose significant challenges to extracting robust spatial features for precise tumor boundary segmentation. The brain processes visual data hierarchically, adapting attention across multiple scales to capture both fine details and broader context simultaneously. Inspired by this, multi-grained information, including local context and long-range dependencies, is introduced to address the challenges in ultrasound image segmentation and improve performance. Existing methods face limitations in this regard: convolutional networks (ConvNets) excel at capturing fine-grained spatial details but fail to account for long-range dependencies, while self-attention mechanisms focus on global context but struggle with continuous local information extraction. The Mamba architecture strikes a balance between local and global information but is not fully optimal in either domain. To address these shortcomings, BMSNet is proposed as a novel deep learning-based architecture that combines the advantages of ConvNets, self-attention mechanisms, and Mamba, aiming to improve the precision of ultrasound image segmentation. Experiments conducted on six ultrasound datasets demonstrate that BMSNet outperforms existing methods across three key metrics: binary accuracy (ACC), intersection over union (IoU), and dice coefficient (DSC).