VcaNet: Vision Transformer with fusion channel and spatial attention module for 3D brain tumor segmentation

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 186, P. 109662 - 109662

Published: Jan. 14, 2025

Language: Английский

---

multiPI-TransBTS: A multi-path learning framework for brain tumor image segmentation based on multi-physical information

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 191, P. 110148 - 110148

Published: April 10, 2025

Language: Английский

---

DBMAE-Net: A dual branch multi-scale feature adaptive extraction network for retinal arteriovenous vessel segmentation

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 104, P. 107619 - 107619

Published: Jan. 29, 2025

Language: Английский

---

Accurate Multimodal Liver Registration of 3D Ultrasound and CT Volume: An Open Dataset and a Model Fusion Method

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 105, P. 107597 - 107597

Published: Feb. 5, 2025

Language: Английский

---

A 3D medical image segmentation network based on gated attention blocks and dual-scale cross-attention mechanism

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 20, 2025

In the field of multi-organ 3D medical image segmentation, Convolutional Neural Networks (CNNs) are limited to extracting local feature information, while Transformer-based architectures suffer from high computational complexity and inadequate extraction spatial channel layer information. Moreover, large number varying sizes organs be segmented result in suboptimal model robustness segmentation outcomes. To address these challenges, this paper introduces a novel network architecture, DS-UNETR++, specifically designed for segmentation. The proposed features dual-branch encoding mechanism that categorizes images into coarse-grained fine-grained types before processing them through blocks. Each block comprises downsampling Gated Shared Weighted Pairwise Attention (G-SWPA) submodule, which dynamically adjusts influence attention on extraction. Additionally, Dual-Scale Cross-Attention Module (G-DSCAM) is incorporated at bottleneck stage. This module employs dimensionality reduction techniques cross-coarse-grained features, using gating balance ratio two thereby achieving effective multi-scale fusion. Finally, comprehensive evaluations were conducted four public datasets. Experimental results demonstrate DS-UNETR++ achieves good performance, highlighting effectiveness significance method offering new insights various organ tasks.

Language: Английский

---

Brain tumor image segmentation based on shuffle transformer-dynamic convolution and inception dilated convolution

Computer Vision and Image Understanding, Journal Year: 2025, Volume and Issue: unknown, P. 104324 - 104324

Published: Feb. 1, 2025

Language: Английский

---

Adjacent slices disentangled 2.5D network for spinal segmentation on multi-view MR images

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 107, P. 107809 - 107809

Published: March 14, 2025

Language: Английский

---

Unsupervised motion artifacts reduction for cone-beam CT via enhanced landmark detection

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 127258 - 127258

Published: March 1, 2025

Language: Английский

---

Medical Image Segmentation with an Emphasis on Prior Convolution and Channel Multi-branch Attention

Digital Signal Processing, Journal Year: 2025, Volume and Issue: unknown, P. 105175 - 105175

Published: March 1, 2025

Language: Английский

---

Multimodal medical image fusion combining saliency perception and generative adversarial network

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 27, 2025

Multimodal medical image fusion is crucial for enhancing diagnostic accuracy by integrating complementary information from different imaging modalities. Current techniques face challenges in effectively combining heterogeneous features while preserving critical information. This paper presents a Temporal Decomposition Network (TDN), novel deep learning architecture that optimizes multimodal through feature-level temporal analysis and adversarial mechanisms. The TDN incorporates two key components: salient perception model discriminative feature extraction generative network matching. identifies classifies distinct pixel distributions across modalities, the component facilitates accurate mapping fusion. approach enables precise of robust quality assessment fused regions. Experimental validation on diverse datasets, encompassing multiple modalities dimensions, demonstrates TDN's superior performance. Compared to state-of-the-art methods, framework achieves an 11.378% improvement 12.441% enhancement precision. These results indicate significant potential clinical applications, particularly radiological diagnosis, surgical planning, analysis, where visualization interpretation decision-making.

Language: Английский

---