Measuring glomerular number from kidney mri images

Patent metadata

Field	Value
Publication number	US-2016005170-A1
Application number	US-201514853645-A
Country	US
Kind code	A1
Filing date	Sep 14, 2015
Priority date	Mar 14, 2013
Publication date	Jan 7, 2016
Grant date	—

Abstract

Measuring the number of glomeruli in the entire, intact kidney using non-destructive techniques is of immense importance in studying several renal and systemic diseases. In particular, a recent Magnetic Resonance Imaging (MRI) technique, based on injection of a contrast agent, cationic ferritin, has been effective in identifying glomerular regions in the kidney. In various embodiments, a low-complexity, high accuracy method for obtaining the glomerular count from such kidney MRI images is described. This method employs a patch-based approach for identifying a low-dimensional embedding that enables the separation of glomeruli regions from the rest. By using only a few images marked by the expert for learning the model, the method provides an accurate estimate of the glomerular number for any kidney image obtained with the contrast agent. In addition, the implementation of our method shows that this method is near real-time, and can process about 5 images per second.

First claim

What is claimed is:

1 . A method of counting glomeruli in an image of a kidney, the method being implemented via execution of computer instructions configured to run at one or more processing modules and configured to be stored at one or more non-transitory memory storage modules, the method comprising:
analyzing patches corresponding to each pixel within the image; determining, using the patches, whether each pixel belongs to at least one of the glomeruli; and counting the glomeruli in the image.

2 . The method of claim 1 , further comprising:
using graph-based embedding to exploit similarities between the patches.

3 . The method of claim 1 , further comprising:
using expert-marked ground truth images to learn a discriminative embedding.

4 . The method of claim 1 , further comprising:
considering relations between glomerular and non-glomerular regions.

5 . The method of claim 1 , further comprising:
adopting a non-local subspace approach.

6 . The method of claim 1 , wherein:
the method is robust to measurement noise in the image.

7 . The method of claim 1 , wherein:
projection directions are pre-computed using training patches.

8 . The method of claim 1 , further comprising:
for each patch in the image, obtaining low dimensional embeddings using an inner product operation.

9 . The method of claim 1 , further comprising:
performing glomeruli identification using K-Means clustering.

10 . The method of claim 1 , further comprising:
displaying at least a portion of the glomeruli on a screen.

11 . The method of claim 1 , wherein:
the graphs are made robust to intensity variations across several images.

12 . The method of claim 11 , wherein:
weights in the graphs are obtained based on sparse representations of the patches.

13 . A method of counting glomeruli in an image of a kidney, the method being implemented via execution of computer instructions configured to run at one or more processing modules and configured to be stored at one or more non-transitory memory storage modules, the method comprising:
extracting a first patch centered around each pixel of the image; computing a low-dimensional projection for each first patch; performing segmentation using K-Means clustering to obtain independent regions; and counting a total number of the independent regions.

14 . The method of claim 13 , wherein:
the first patches are each 5 pixels by 5 pixels.

15 . The method of claim 13 , further comprising:
normalizing each first patch.

16 . The method of claim 13 , wherein:
computing the low-dimensional projection for each first patch is based at least in part on a projection matrix obtained from training images.

17 . The method of claim 13 , further comprising:
extracting a second patch centered around each pixel of training images; identifying the second patches that are centered at positive pixels and the second patches that are centered at negative pixels using expert-marked ground truth images; constructing l 1 graphs to model inter-class and intra-class relationships; and performing local discriminant embedding.

18 . The method of claim 17 , further comprising:
normalizing each second patch centered around each pixel of the training images.

19 . A system for counting glomeruli in an image of a kidney, the system comprising:
one or more processing modules; and one or more non-transitory memory storage modules storing computing instructions configured to run on the one or more processing modules and perform the acts of:
extracting a first patch centered around each pixel of the image;
computing a low-dimensional projection for each first patch;
performing segmentation using K-Means clustering to obtain independent regions; and
counting a total number of the independent regions.

20 . The system of claim 19 , wherein the computing instruction are further configured to perform the acts of:
extracting a second patch centered around each pixel of training images; identifying the second patches that are centered at positive pixels and the second patches that are centered at negative pixels using expert-marked ground truth images; constructing l 1 graphs to model inter-class and intra-class relationships; and performing local discriminant embedding.

Assignees

• Thiagarajan Jayaraman Jayaraman
• Ramamurthy Karthikeyan
• Spanias Andreas
• Frakes David
• Univ Arizona

Inventors

• Thiagarajan Jayaraman Jayaraman
• Ramamurthy Karthikeyan
• Spanias Andreas
• Frakes David

Classifications

• A61P13/12

  of the kidneys · CPC title

• G06T11/60

  Creating or editing images; Combining images with text · CPC title

• A61B5/1075

  for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer (A61B8/0858 takes precedence) · CPC title

• G06T2207/10004

  Still image; Photographic image · CPC title

• A61B5/20

  for measuring urological functions {restricted to the evaluation of the urinary system (A61B5/4375 takes precedence)} · CPC title