Aug 2026· IEEE journal of biomedical and health informatics· Vol PP, pp. 1-14· 0 citations
Medicine
TL;DR
A hierarchical disentanglement guided diffusion framework (HD-Diff) for multimodal brain tumor segmentation is proposed, and a modality-aware en coder is introduced to harmonize and integrate multimodal features.
Abstract
Brain tumor segmentation from magnetic resonance imaging (MRI) is crucial for diagnosis and treatment. However, it faces the challenge of multi-level se mantic perception. The challenge includes local modality specificity and global subregion heterogeneity. Diffusion based segmentation methods can potentially address this challenge, but they are hindered by two critical issues: in adequate modeling of modality interactions and inappropri ate coupling of structural conditions with denoising stages. In this paper, we propose a hierarchical disentanglement guided diffusion framework (HD-Diff) for multimodal brain tumor segmentation. We introduce a modality-aware en coder to harmonize and integrate multimodal features. A dual-stream feature fusion module is employed to integrate extracted multimodal features across multiple scales. This module eliminates semantic misalignment and spatial in consistencies between modalities caused by differences in imaging principles. We utilize boundary and core enhanced conditioners to explicitly encode structural prior information about the tumor boundary and core regions. Thenboundaryand corefeatures are hierarchically injected into the denoising process. This strategy achieves independent optimization of both the overall tumor topology and local edge details. Extensive experiments are conducted on multiple public brain tumor segmentation datasets. Across these five benchmarks, HD-Diff achieves strong overallseg mentation performance, with highly competitive Dice and HD95 results. The consistent results further demonstrate its effectiveness across diverse brain tumor datasets.
Experimental results on the BraTS 2021 and BraTS-Africa datasets demonstrate that DuS-Net achieves higher segmentation accuracy than recent mainstream segmentation methods, validating its effectiveness for complex brain glioma segmentation and indicating its potential value for computer-assisted glioma segmentation.
Zhe Du, Tian Tang, Kepeng Yang et al.· IEEE Access· 0 citations
The segmentation of brain tumors in multimodal magnetic resonance imaging (MRI) is an important task in the clinical diagnosis, treatment planning and disease monitoring processes. The accurate delineation is however difficult because of the heterogeneity of tumors, irregularity of their borders and the different inten...
Rashmi Ashtagi· Natural Resources for Human...· 0 citations
A 3D U-Net-based segmentation framework, NeuroFocusNet, which synchronizes spatial attention, channel recalibration, and attention-gated skip connections within a single volumetric encoder-decoder model is proposed, which is encouraging when considering the feasibility of using NeuroFocusNet as a technically promising...
Faizan Ullah, Z. Abbas, Sergo Gegechkori et al.· IEEE Access· 0 citations
Evaluation on the BraTS GOAT validation dataset demonstrated that the proposed Multi-Stage Dynamic Prompt nnU-Net outperformed the baseline nnU Net across the majority of evaluated metrics and tumor subregions.
Mohammad Mahdi Danesh Pajouh, Sara Saeedi· 0 citations
Abstract This study proposes a 2.5D input framework for brain tumor segmentation in multi-sequence magnetic resonance imaging (MRI), aiming to balance computational efficiency and spatial context compared to conventional 2D approaches. While 2D methods are efficient but limited in capturing inter-slice information, the...
Sawehel Tlahig, Max Bindemann, T. Schanze· Current Directions in Biomed...· 0 citations
OW-Mamba is a modality adaptive 3D segmentation framework for accurate and stable single modality inference while supporting arbitrary valid modality combinations, and introduces a modality adaptive base residual fusion strategy for stable feature integration with controlled local refinement.