Image-based detection of the boundary between a first part and a second part

Patent metadata

Field	Value
Publication number	US-9582729-B2
Application number	US-201314443203-A
Country	US
Kind code	B2
Filing date	Nov 4, 2013
Priority date	Nov 30, 2012
Publication date	Feb 28, 2017
Grant date	Feb 28, 2017

Abstract

A detection apparatus for detecting the position of a boundary between a first part and a second part of a subject, includes a pixel extraction unit for extracting a plurality of candidate pixels acting as candidates for a pixel situated on the boundary on the basis of image data of a first section crossing the first part and the second part, and a pixel specification unit for specifying the pixel situated on the boundary from within the plurality of candidate pixels by using an identifier which has been prepared by using an algorithm of machine learning.

First claim

The invention claimed is:

1. A detection apparatus for detecting the position of a boundary between a first part and a second part of a subject, the detection apparatus comprising:
a pixel extraction unit for extracting a plurality of candidate pixels acting as candidates for a pixel situated on the boundary on the basis of image data of a first section crossing the first part and the second part; and
a pixel specification unit for specifying the pixel situated on the boundary from within the plurality of candidate pixels by using an identifier which has been prepared by using an algorithm of machine learning;
wherein the pixel specification unit narrows down candidate pixels which are high in possibility that they are situated on the boundary from within the plurality of candidate pixels, and specifies the pixel which is situated on the boundary from within the narrowed-down candidate pixels by using the identifier.

2. The detection apparatus according to claim 1 , wherein the identifier is prepared by making it learn supervised data by AdaBoost.

3. The detection apparatus according to claim 1 , wherein the identifier is prepared by making it learn supervised data by Support Vector Machine.

4. The detection apparatus according to claim 1 , further comprising a navigator region determination unit for determining the position of the navigator region on the basis of the specified pixel.

5. The detection apparatus according to claim 4 ,
wherein the pixel extraction unit extracts the plurality of candidate pixels per the first section on the basis of image data of a plurality of first sections crossing the first part and the second part;
the pixel specification unit specifies a set of pixels situated on the boundary per the first section; and
the navigator region determination unit selects a set of pixels to be used for determining the position of the navigator region from within the sets of pixels specified per the first section, and determines the position of the navigator region on the basis of the selected set of pixels.

6. The detection apparatus according to claim 5 , wherein the navigator region determination unit decides whether there exists a gap of pixels in the selected set of pixels, when there exists the gap of pixels, bridges the gap of pixels and determines the position of the navigator region on the bases of the set of pixels after the gap of pixels has been bridged.

7. The detection apparatus according to claim 6 , wherein the navigator region determination unit performs a fitting process on the set of pixels after the gap of pixels has been bridged and determines the position of the navigator region on the basis of the set of pixels after fitting-processed.

8. The detection apparatus according to claim 7 , wherein the navigator region determination unit determines a pixel situated at the uppermost position in the set of pixels after fitting-processed as the position of the navigator region.

9. The detection apparatus according to claim 1 , wherein image data of the first section is differentiated image data.

10. The detection apparatus according to claim 9 , wherein the pixel extraction unit obtains a profile of differential values of pixels on a line crossing the boundary using the differentiated image data and extracts the candidate pixels on the basis of the profile.

11. The detection apparatus according to claim 10 , wherein the pixel specification unit narrows down pixels which are high in possibility that they are situated on the boundary from within the two or more candidate pixels on the basis of pixel values of pixels situated around each candidate pixel on the line, in a case where two or more candidate pixels have been extracted on the line.

12. The detection apparatus according to claim 11 , wherein the pixel specification unit sets a first region and a second region for the candidate pixels, and narrows down pixels which are high in possibility that they are situated on the boundary on the basis of pixel values of pixels included in the first region and pixel values of pixels included in the second region.

13. The detection apparatus according to claim 1 , wherein the pixel extraction unit obtains a search region including the boundary, and extracts the candidate pixels from within the search region.

14. The detection apparatus according to claim 13 , wherein in a case where search region is to be obtained, image data of a second section intersecting with the first section is used.

15. The detection apparatus according to claim 14 , wherein the first section is a coronal plane and the second section is an axial plane.

16. The detection apparatus according to claim 1 , wherein the image data is image data that fat has been removed.

17. The detection apparatus according to claim 1 , wherein the first part is the lung and the second part is the liver.

18. A magnetic resonance apparatus for detecting the position of a boundary between a first part and a second part of a subject, the magnetic resonance apparatus comprising:
a pixel extraction unit for extracting a plurality of candidate pixels acting as candidates for a pixel situated on the boundary on the basis of image data of a first section crossing the first part and the second part; and
a pixel specification unit for specifying the pixel situated on the boundary from within the plurality of candidate pixels by using an identifier which has been prepared by using an algorithm of machine learning;
wherein the pixel specification unit narrows down candidate pixels which are high in possibility that they are situated on the boundary from within the plurality of candidate pixels, and specifies the pixel which is situated on the boundary from within the narrowed-down candidate pixels by using the identifier.

19. A detection method of detecting the position of a boundary between a first part and a second part of a subject, the detection method comprising:
the pixel extraction step of extracting a plurality of candidate pixels acting as candidates for a pixel situated on the boundary on the basis of image data of a first section crossing the first part and the second part; and
the pixel specification step of specifying the pixel situated on the boundary from within the plurality of candidate pixels by using an identifier which has been prepared by using an algorithm of machine learning;
wherein the pixel specification step narrows down candidate pixels which are high in possibility that they are situated on the boundary from within the plurality of candidate pixels, and specifies the pixel which is situated on the boundary from within the narrowed-down candidate pixels by using the identifier.

Assignees

• Ge Medical Sys Global Tech Co Llc
• Gen Electric

Inventors

• Goto Takao

Classifications

• G01R33/543 Primary

  Control of the operation of the MR system, e.g. setting of acquisition parameters prior to or during MR data acquisition, dynamic shimming, use of one or more scout images for scan plane prescription (G01R33/546 takes precedence) · CPC title

• G06F18/2413

  based on distances to training or reference patterns · CPC title

• G06T2207/30056

  Liver; Hepatic · CPC title

• G01R33/5608

  Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels (image data processing or generation, in general G06T) · CPC title

• G01R33/5676

  Gating or triggering based on an MR signal, e.g. involving one or more navigator echoes for motion monitoring and correction · CPC title