Granted Patent B2

US 12,373,951 · App. 18/744,328 · Granted Jul 29, 2025

Inventors: Michael Edward Sughrue (Sydney, AU); Stephane Philippe Doyen (Glebe, AU); Charles Teo (Sydney, AU)

Assignee: Omniscient Neurotechnology Pty Limited

G06T7/0014G01R33/5608G01R33/56341G06T7/30G06T2207/10088G06T2207/20076G06T2207/20081G06T2207/30016

Abstract

Claims (52)

1. A method comprising:

obtaining brain tractography data of a particular brain;

registering the brain tractography data for the particular brain in a three-dimensional coordinate system space;

registering a standard atlas in the three-dimensional coordinate system space to produce a registered atlas, the registered atlas containing a parcellation scheme;

for at least one voxel and a first parcellation in the registered atlas:

determining end-to-end voxel level tractography vectors showing end-to-end connectivity of the voxel with voxels in other parcellations;

classifying the voxel based on probability of the voxel being part of the first parcellation; and

repeating the determining of the end-to-end voxel level tractography vectors and the classifying of the voxels for a plurality of parcellations of the registered atlas to form a personalised brain atlas containing an adjusted parcellation scheme reflecting the particular brain.

2. The method according to claim 1 , further comprising, prior to the repeating step, the step of interpolating a voxel grid for end-to-end parcellated voxel level tractography vectors to fill gaps between voxels.

3. The method according to claim 2 , further comprising, for each voxel in a particular parcellation in the registered atlas:

determining end-to-end voxel level tractography vectors and pass-through parcellated voxel level tractography vectors respectively showing end-to-end and pass-through connectivity of the voxel with voxels in other parcellations;

classifying the voxel to determine labels and a voxel grid for end-to-end parcellated voxel level tractography vectors and pass-through parcellated voxel level tractography vectors, based on the probability of the voxel being part of the particular parcellation; and

interpolating the voxel grid for end-to-end parcellated voxel level tractography vectors and pass-through parcellated voxel level tractography vectors to fill gaps between voxels.

4. The method according to claim 1 , wherein the method further comprises determining a registration function for the particular brain in the three dimensional coordinate system space and wherein determining the registration function comprises the steps of:

performing face stripping, skull stripping and masking of a NIfTI version of T1 images of the brain to obtain a masked, skull and face stripped T1 image; and

determining a relationship between the masked, skull and face stripped T1 image and a set of standard brain data image sets to generate the registration function.

5. The method according to claim 1 , wherein the step of determining the registered atlas comprises applying a registration function to the standard Atlas to generate the registered atlas.

6. The method according to claim 1 , wherein obtaining brain tractography data comprises performing diffusion tractography on brain image data using a face stripped masked NIfTI version of DTI images of a DICOM image set.

7. The method according to claim 1 , wherein the determining of the voxel level tractography vectors comprises the steps of:

registering the registered Atlas and a brain tractography image set;

generating end-to-end parcellated voxel level tractography vectors; and

generating end-to-end and pass-through parcellated voxel level tractography vectors.

8. The method according to claim 1 , wherein classifying the voxel comprises processing the end-to-end parcellated voxel level tractography vectors and the pass-through parcellated voxel level tractography vectors with an end-to-end classifier and a pass-by classifier to form a voxel grid.

9. The method according to claim 1 , wherein the plurality of parcellations comprises all parcellations in the registered atlas.

10. The method according to claim 1 , wherein the three dimensional coordinate system space is Montreal Neurological Institute space described by a set (HCP-SDB) of standard brain data image sets.

11. The method according to claim 1 , wherein the standard Atlas is a HCP-MMP1 Atlas.

12. The method of claim 1 , wherein the brain tractography data of the particular brain are whole brain tractography images of the particular brain.

13. A system comprising:

one or more computers and one or more storage devices on which are stored instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

obtaining brain tractography data of a particular brain;

registering the brain tractography data for the particular brain in a specified space;

registering a standard atlas in the specified space to produce a registered atlas, the registered atlas containing a parcellation scheme;

for at least one voxel and a first parcellation in the registered atlas:

determining end-to-end voxel level tractography vectors showing end-to-end connectivity of the voxel with voxels in other parcellations;

classifying the voxel based on the probability of the voxel being part of the particular parcellation; and

repeating the determining of the end-to-end voxel level tractography vectors and the classifying of the voxels for a plurality of parcellations of the registered atlas to form a personalised brain atlas containing an adjusted parcellation scheme reflecting the particular brain.

14. The system according to claim 13 , wherein the operations further comprise, prior to the repeating step, the step of interpolating a voxel grid for end-to-end parcellated voxel level tractography vectors to fill gaps between voxels.

15. The system according to claim 13 , wherein the step of determining the registered atlas comprises applying a registration function to the standard atlas to generate the registered atlas.

16. The system according to claim 13 , wherein the operations further comprises performing diffusion tractography on brain image data, performed in relation to a face stripped masked NIfTI version of DTI images of a DICOM image set.

17. The system according to claim 13 , wherein the plurality of parcellations comprises all parcellations in the registered atlas.

18. The method according to claim 1 , wherein the determining of the voxel level tractography vectors comprises the steps of:

registering the registered Atlas and the brain tractography data;

generating end-to-end parcellated voxel level tractography vectors; and

generating end-to-end and pass-through parcellated voxel level tractography vectors.

19. A computer readable storage medium having one or more computer programs recorded therein, the one or more programs being executable by a computer apparatus to make the computer apparatus perform a method of processing brain image data of a particular brain to be parcellated, the method comprising the steps of:

obtaining brain tractography data of a particular brain;

registering the brain tractography data for the particular brain in a three-dimensional coordinate system space;

registering a standard atlas in the three-dimensional coordinate system space to produce a registered atlas, the registered atlas containing a parcellation scheme;

for at least one voxel and a first parcellation in the registered atlas:

determining end-to-end voxel level tractography vectors showing end-to-end connectivity of the voxel with voxels in other parcellations;

classifying the voxel based on the probability of the voxel being part of the particular parcellation; and

repeating the determining of the end-to-end voxel level tractography vectors and the classifying of the voxels for a plurality of parcellations of the registered atlas to form a personalised brain atlas containing an adjusted parcellation scheme reflecting the particular brain.

Assignments (4)

Priority Claims (1)

Continuity (3)

References Cited (56)