Fiber tracking and segmentation
The present solution can segment tracts by performing two-pass tractography. The system can first perform deterministic tractography and then probabilistic tractography. The system can use the result from the deterministic tractography to update and refine initial identified regions of interest. The refined regions of interest can be used to filter and select streamlines identified through the probabilistic tractography.
1. A data processing system comprising one or more processors to segment neurological tracts, the data processing system configured to:
generate, based on a diffusion-weighted (DW) image of image data, a first plurality of streamlines indicating a fiber tract in the DW image;
select a subset of a first plurality of voxels associated with an anatomical image of the image data based on at least one of the first plurality of streamlines passing through each voxel of the subset of the first plurality of voxels, wherein the subset of the first plurality of voxels is associated with a region of the anatomical image;
generate a second plurality of streamlines, each of the second plurality of streamlines indicating a candidate fiber tract;
select a subset of the second plurality of streamlines, each streamline of the subset of the second plurality of streamlines passes through the region of the anatomical image; and
generate a tract image comprising the subset of the second plurality of streamlines, wherein each streamline of the subset of the second plurality of streamlines passes through the region of the anatomical image.
2. The data processing system of claim 1 , wherein the data processing system generates the first plurality of streamlines indicating the fiber tract with deterministic tractography.
3. The data processing system of claim 1 , wherein the data processing system generates the second plurality of streamlines with probabilistic tractography.
4. The data processing system of claim 1 , wherein the region is a part of a plurality of regions of the anatomical image, and wherein the data processing system maps the plurality of regions from a template to the anatomical image.
5. The data processing system of claim 4 , wherein the template comprises a Montreal Neurological Institute (MNI) template image.
6. The data processing system of claim 4 , wherein the data processing system warps the template to the anatomical image with a symmetric, invertible warp.
7. The data processing system of claim 1 , wherein the data processing system generates the first plurality of streamlines using constrained spherical deconvolution.
8. The data processing system of claim 1 , wherein the tract image comprises the subset of the second plurality of streamlines aligned with the anatomical image.
9. The data processing system of claim 1 , wherein the anatomical image is an MRI image.
10. The data processing system of claim 1 , wherein the data processing system is configured to:
select a subset of a second plurality of voxels associated with the anatomical image of the image data based on at least one of the first plurality of streamlines passing through each voxel of the subset of the second plurality of voxels, wherein the subset of the second plurality of voxels is associated with a second region of the anatomical image; and
select the subset of the second plurality of streamlines, wherein each streamline of the subset of the second plurality of streamlines passes through the second region.
11. A method to segment neurological tracts, comprising:
generating, by a segmentation engine, based on a diffusion-weighted (DW) image of image data, a first plurality of streamlines indicating a fiber tract in the DW image;
selecting, by the segmentation engine, a subset of a first plurality of voxels associated with an anatomical image of the image data based on at least one of the first plurality of streamlines passing through each voxel of the subset of the first plurality of voxels, wherein the subset of the first plurality of voxels is associated with a region of the anatomical image,
generating, by the segmentation engine, a second plurality of streamlines, each of the second plurality of streamlines indicating a candidate fiber tract;
selecting, by the segmentation engine, a subset of the second plurality of streamlines, each streamline of the subset of the second plurality of streamlines passes through the region of the anatomical image; and
generating, by the segmentation engine, a tract image comprising the subset of the second plurality of streamlines, wherein each streamline of the subset of the second plurality of streamlines passes through the region of the anatomical image.
12. The method of claim 11 , further comprising generating the first plurality of streamlines indicating the fiber tract with deterministic tractography.
13. The method of claim 11 , further comprising generating the second plurality of streamlines with probabilistic tractography.
14. The method of claim 11 , wherein the region is a part of a plurality of regions of the anatomical image, further comprising mapping the plurality of regions from a template to the anatomical image.
15. The method of claim 14 , wherein the template comprises a Montreal Neurological Institute (MNI) template image.
16. The method of claim 14 , further comprising warping the template to the anatomical image with a symmetric, invertible warp.
17. The method of claim 11 , further comprising generating the first plurality of streamlines using constrained spherical deconvolution.
18. The method of claim 11 , wherein the tract image comprises the subset of the second plurality of streamlines aligned with the anatomical image.
19. The method of claim 11 , wherein the anatomical image is an MRI image.
20. The method of claim 11 , further comprising:
selecting, by the segmentation engine, a subset of a second plurality of voxels associated with the anatomical image of the image data based on at least one of the first plurality of streamlines passing through each voxel of the subset of the second plurality of voxels, wherein the subset of the second plurality of voxels is associated with a second region of the anatomical image; and
selecting, by the segmentation engine, the subset of the second plurality of streamlines, wherein each streamline of the subset of the second plurality of streamlines passes through the second region.