Deep Learning-Based Diffusion MRI Tractography: Integrating Spatial and Anatomical Information
This work addresses the challenge of false-positive connections in brain white matter pathway reconstruction for neuroscience and clinical applications, representing an incremental improvement over existing deep learning methods.
The paper tackled the problem of inaccurate diffusion MRI tractography by introducing a deep learning framework that integrates spatial and anatomical information, achieving a valid streamline rate of 66.2% and white matter coverage of 63.8% on a simulated dataset, with improvements in coverage and reduction in overreach on a multi-site dataset.
Diffusion MRI tractography technique enables non-invasive visualization of the white matter pathways in the brain. It plays a crucial role in neuroscience and clinical fields by facilitating the study of brain connectivity and neurological disorders. However, the accuracy of reconstructed tractograms has been a longstanding challenge. Recently, deep learning methods have been applied to improve tractograms for better white matter coverage, but often comes at the expense of generating excessive false-positive connections. This is largely due to their reliance on local information to predict long range streamlines. To improve the accuracy of streamline propagation predictions, we introduce a novel deep learning framework that integrates image-domain spatial information and anatomical information along tracts, with the former extracted through convolutional layers and the later modeled via a Transformer-decoder. Additionally, we employ a weighted loss function to address fiber class imbalance encountered during training. We evaluate the proposed method on the simulated ISMRM 2015 Tractography Challenge dataset, achieving a valid streamline rate of 66.2%, white matter coverage of 63.8%, and successfully reconstructing 24 out of 25 bundles. Furthermore, on the multi-site Tractoinferno dataset, the proposed method demonstrates its ability to handle various diffusion MRI acquisition schemes, achieving a 5.7% increase in white matter coverage and a 4.1% decrease in overreach compared to RNN-based methods.