Technologies for diagnosing neurological or psychiatric illnesses

Patent metadata

Field	Value
Publication number	US-2017039714-A1
Application number	US-201515304611-A
Country	US
Kind code	A1
Filing date	Apr 17, 2015
Priority date	Apr 17, 2014
Publication date	Feb 9, 2017
Grant date	—

Abstract

A technology which enables identifying, via a computer, a vessel in a third image. The third image is obtained from a subtraction of a second image from a first image. The second image and the first image are aligned on an imaging space. The first image is post-contrast. The second image is pre-contrast. The technology enables determining, via the computer, a voxel intensity mean value of a segment of the vessel in the third image. The technology enables obtaining, via the computer, a fourth image from a division of the third image by the voxel intensity mean value. The technology enables applying, via the computer, a filter onto the fourth image. The technology enables generating, via the computer, a filter mask based on the fourth image.

First claim

1 . A method comprising:
identifying, via a computer, a vessel in a third image, wherein the third image is obtained from a subtraction of a second image from a first image, wherein the second image and the first image are aligned within an imaging space, wherein the first image is post-contrast, wherein the second image is pre-contrast; determining, via the computer, a voxel intensity mean value of a segment of the vessel in the third image; obtaining, via the computer, a fourth image from a division of the third image by the voxel intensity mean value; applying, via the computer, a filter onto the fourth image; and generating, via the computer, a filter mask based on the fourth image.

2 . The method of claim 1 , further comprising:
performing, via the computer, a vessel segmentation process on at least one of the first image or the second image before the identifying.

3 . The method of claim 2 , wherein the performing is automatically triggered.

4 . The method of claim 1 , wherein the vessel comprises a diameter of about one centimeter or less.

5 . The method of claim 1 , wherein the identifying is based on a vesselness filter and a pre-defined region of interest, wherein the vesselness filter filters based on a set of eigenvalues of a Hessian matrix of the third image, wherein the third image is modified such that the region of interest is positioned in a predefined area.

6 . The method of claim 1 , wherein the voxel intensity mean value is based on a highest voxel intensity range in the segment, wherein the range comprises a top 40% of voxel intensities.

7 . The method of claim 6 , wherein the range comprises a top 33% of voxel intensities.

8 . The method of claim 6 , wherein the range comprises a top 25% of voxel intensities.

9 . The method of claim 1 , wherein the third image is a cerebral blood volume map, wherein the filter is based on at least one of a performance of an expectation-maximization segmentation, or a fitting of a bimodal Gaussian curve to a histogram of data in accordance with the third image.

10 . The method of claim 1 , wherein the filter mask is a binary mask, and further comprising:
applying, via the computer, the binary mask to the third image; and mapping, via the computer, based on the applying, the third image according to a change in a transverse relaxation time induced via an input of a contrast agent.

11 . A system comprising:
a hardware processor; a memory coupled to the hardware processor, wherein the memory stores a set of instructions to execute via the hardware processor, wherein the instructions instruct the hardware processor to perform a method comprising:
identifying, via a computer, a vessel in a third image, wherein the third image is obtained from a subtraction of a second image from a first image, wherein the second image and the first image are aligned within an imaging space, wherein the first image is post-contrast, wherein the second image is pre-contrast;
determining, via the computer, a voxel intensity mean value of a segment of the vessel in the third image;
obtaining, via the computer, a fourth image from a division of the third image by the voxel intensity mean value;
applying, via the computer, a filter onto the fourth image; and
generating, via the computer, a filter mask based on the fourth image.

12 . The system of claim 11 , wherein the method further comprises:
performing, via the computer, a vessel segmentation process on at least one of the first image or the second image before the identifying.

13 . The system of claim 12 , wherein the performing is automatically triggered.

14 . The system of claim 11 , wherein the vessel comprises a diameter of about one centimeter or less.

15 . The system of claim 11 , wherein the identifying is based on a vesselness filter and a pre-defined region of interest, wherein the vesselness filter filters based on a set of eigenvalues of a Hessian matrix of the third image, wherein the third image is modified such that the region of interest is positioned in a predefined area.

16 . The system of claim 11 , wherein the voxel intensity mean value is based on a highest voxel intensity range in the segment, wherein the range comprises a top 40% of voxel intensities.

17 . The system of claim 16 , wherein the range comprises a top 33% of voxel intensities.

18 . The system of claim 16 , wherein the range comprises a top 25% of voxel intensities.

19 . The system of claim 11 , wherein the third image is a cerebral blood volume map, wherein the filter is based on at least one of a performance of an expectation-maximization segmentation, or a fitting of a bimodal Gaussian curve to a histogram of data in accordance with the third image.

20 . The system of claim 11 , wherein the filter mask is a binary mask, and wherein the method further comprises:
applying, via the computer, the binary mask to the third image; and mapping, via the computer, based on the applying, the third image according to a change in a transverse relaxation time induced via an input of a contrasting agent.

21 . A computer-readable storage device storing a set of instructions for execution via a processing circuit to implement a method, wherein the method comprises:
identifying, via a computer, a vessel in a third image, wherein the third image is obtained from a subtraction of a second image from a first image, wherein the second image and the first image are aligned within an imaging space, wherein the first image is post-contrast, wherein the second image is pre-contrast; determining, via the computer, a voxel intensity mean value of a segment of the vessel in the third image; obtaining, via the computer, a fourth image from a division of the third image by the voxel intensity mean value; applying, via the computer, a filter onto the fourth image; and generating, via the computer, a filter mask based on the fourth image.

22 . The computer-readable storage device of claim 21 , wherein the method further comprises:
performing, via the computer, a vessel segmentation process on at least one of the first image or the second image before the identifying.

23 . The computer-readable storage device of claim 22 , wherein the performing is automatically triggered.

24 . The computer-readable storage device of claim 21 , wherein the vessel comprises a diameter of about one centimeter or less.

25 . The computer-readable storage device of claim 21 , wherein the identifying is based on a vesselness filter and a pre-defined region of interest, wherein the vesselness filter filters based on a set of eigenvalues of a Hessian matrix of the third image, wherein the third image is modified such that the region of interest is positioned in a predefined area.

26 . The computer-readable storage device of claim 21 , wherein the voxel intensity mean value is based on a highest voxel intensity range in the segment, wherein the range comprises a top 40% of voxel intensities.

27 . The computer-readable storage device of claim 26 , wherein the range comprises a top 33% of voxel intensities.

28 . The computer-readable storage device of claim 26 , wherein the range comprises a top 25% of voxel intensities.

29 . The computer-readable storage device of claim 21 , wherein the third image is a cerebral blood volume map, wherein the filter is base

Assignees

• Univ Columbia

Inventors

• Small Scott A
• Provenzano Frank A
• Khan Usman A

Classifications

• G06T2207/20224

  Image subtraction · CPC title

• G06T7/0016 Primary

  involving temporal comparison · CPC title

• G06T2207/30016

  Brain · CPC title

• G06T5/20

  using local operators · CPC title

• G06T2207/10088

  Magnetic resonance imaging [MRI] · CPC title