Technology Category
Computer Vision: Diffusion Models for VisionMachine Learning: Deep Neural Architectures and Foundation ModelsCognitive Modeling & Cognitive Systems: Neural Spike Coding
TractoTransformer: Diffusion MRI Streamline Tractography using CNN and Transformer Networks
🤖 AI Summary
Diffusion MRI data suffer from high noise levels and low spatial resolution, leading to frequent failures of white matter fiber tractography—particularly in crossing, merging, and fanning regions. To address this, we propose a hybrid CNN-Transformer architecture: convolutional neural networks extract local diffusion microstructural features, while the Transformer captures long-range sequential dependencies along fiber trajectories, enabling context-aware, end-to-end fiber orientation prediction and tractography. This work is the first to synergistically integrate CNNs and Transformers for multi-scale fiber modeling. Our method significantly improves path completeness and angular accuracy in anatomically complex regions. Evaluated on the Tractometer benchmark, it achieves state-of-the-art performance; on the real-world clinical dataset TractoInferno, it demonstrates strong generalizability—improving path completeness by 12.7% and reducing mean angular error by 23.4% compared to prior methods.
Technology Category
Application Category
📝 Abstract
White matter tractography is an advanced neuroimaging technique that reconstructs the 3D white matter pathways of the brain from diffusion MRI data. It can be framed as a pathfinding problem aiming to infer neural fiber trajectories from noisy and ambiguous measurements, facing challenges such as crossing, merging, and fanning white-matter configurations. In this paper, we propose a novel tractography method that leverages Transformers to model the sequential nature of white matter streamlines, enabling the prediction of fiber directions by integrating both the trajectory context and current diffusion MRI measurements. To incorporate spatial information, we utilize CNNs that extract microstructural features from local neighborhoods around each voxel. By combining these complementary sources of information, our approach improves the precision and completeness of neural pathway mapping compared to traditional tractography models. We evaluate our method with the Tractometer toolkit, achieving competitive performance against state-of-the-art approaches, and present qualitative results on the TractoInferno dataset, demonstrating strong generalization to real-world data.
Problem
Research questions and friction points this paper is trying to address.
Reconstructing brain white matter pathways from noisy diffusion MRI data
Modeling sequential streamline trajectories using Transformer and CNN networks
Improving precision and completeness of neural fiber tractography mapping
Innovation
Methods, ideas, or system contributions that make the work stand out.
Leveraging Transformers to model streamline sequential nature
Utilizing CNNs to extract microstructural features locally
Combining Transformer and CNN for improved pathway mapping
🔎 Similar Papers
No similar papers found.
