Related Experiment Video
Updated: Aug 7, 2025
17:06
Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging
Published on: November 8, 2012
26.3K
Convolutional-recurrent neural networks approximate diffusion tractography from T1-weighted MRI and associated
Leon Y Cai1, Ho Hin Lee2, Nancy R Newlin2
1Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA.
Biorxiv : the Preprint Server for Biology
|March 13, 2023
Summary
This study introduces a deep learning method to perform brain white matter tractography using T1w MRI, bypassing the need for time-consuming diffusion MRI. This approach achieves accuracy comparable to traditional diffusion MRI methods.
Area of Science:
- Neuroimaging
- Computational Neuroscience
- Artificial Intelligence
Background:
- Diffusion MRI (dMRI) streamline tractography is essential for mapping brain white matter (WM) pathways.
- Clinical dMRI acquisitions often lack the high angular resolution required for detailed tractography, limiting its use.
- Deep learning models can learn streamline propagation directly from dMRI data.
Conclusions:
- Tractography can be performed using T1w MRI with deep learning, offering a viable alternative to dMRI.
- The findings challenge the necessity of microstructural information for tractography, suggesting dMRI may have primarily enabled its implementation.
- This method expands the potential for routine clinical tractography analyses.
Keywords:
T1-weighted MRIbundlesconnectomicsconvolutional-recurrent neural networksdiffusion MRItractographywhite matter
