Method and device for segmenting a medical examination object with quantitative magnetic resonance imaging

Patent metadata

Field	Value
Publication number	US-2016155238-A1
Application number	US-201514956755-A
Country	US
Kind code	A1
Filing date	Dec 2, 2015
Priority date	Dec 2, 2014
Publication date	Jun 2, 2016
Grant date	—

Abstract

In a method and magnetic resonance apparatus for segmenting image data of an examination object, raw data of the examination object are achieved with the operation of a magnetic resonance scanner. Quantitative image data of the examination object are then calculated in a processor from the raw data. At least one physical variable of the examination object is quantitatively ascertained pixelwise and is displayed. The quantitative image data are segmented for identification of predetermined objects in the quantitative image data, and displayed in a display unit.

First claim

We claim as our invention:

1 . A method for segmenting image data of an examination object comprising:
providing raw data to a computer, said raw data having been acquired from an examination object exhibiting at least one physical variable; in said computer, automatically determining quantitative image data of the examination object from the raw data and displaying said quantitative raw image data at a display screen in communication with said computer as a plurality of pixels with each pixel presenting a value of said at least one physical variable; and segmenting the displayed quantitative image data to select at least one segment in the displayed quantitative image data that represents a predetermined image object in said quantitative image data.

2 . A method as claimed in claim 1 comprising:
in said computer, determining first quantitative image data from the raw data comprising pixels in which a value of a first physical variable is quantitatively ascertained, and displaying said pixels representing said first physical variable at said display screen;
segmenting said first quantitative image data at said display screen;
determining a first quality parameter for the segmented first image data that designates a degree of reliability that said predetermined object can be identified in the segmented first image data;
in said computer, determining second quantitative image data from the raw data comprising pixels in which a value of a second physical variable is quantitatively ascertained, and displaying said pixels representing said second physical variable at said display screen;
segmenting said second quantitative image data at said display screen;
determining a second quality parameter for the segmented second image data that designates a degree of reliability that said predetermined object can be identified in the segmented second image data; and
in said computer comparing the segmented first quantitative image data with the segmented second quantitative image data and, in any region in which the predetermined object in the segmented first quantitative image data does not match the predetermined object in the segmented second quantitative image data, automatically selecting, as said predetermined object, the quantitative image data having a higher quality parameter associated therewith.

3 . A method as claimed in claim 2 comprising, in said computer, determining an overlaid variable from said first and second physical variable of the examination object.

4 . A method as claimed in claim 1 comprising comparing said physical variable in the quantitative image data of the predetermined object with a first reference database that comprises a plurality of sets of reference data respectively for different examination objects, and a when a predetermined deviation of said physical variable from one of said reference data sets is exceeded, displaying a designation at said display screen that said deviation has been exceeded.

5 . A method as claimed in claim 4 comprising, after segmenting the object in the quantitative image data, further segmenting said object by MR fingerprinting sub-voxel quantification, using at least one physical variable of the pixels from said first reference database.

6 . A method as claimed in claim 5 comprising determining said predetermined object by a procedure selected from the group consisting of thresholding, correlation analysis, region growing, image processing with regard to active objects, and statistical testing, of at least two physical variables in said quantitative image data by comparison with said first reference database.

7 . A method as claimed in claim 6 comprising, in said computer, determining a volume of said predetermined object from said physical variable in the quantitative image data, and comparing said volume with a second reference database that comprises a plurality of sets of volume data respectively for different examination objects, and when a predetermined deviation of said volume from one of said volume data sets is exceeded, displaying a designation at said display screen that said deviation has been exceeded.

8 . A method as claimed in claim 7 comprising supplying said quantitative data to at least one of said first reference database and said second reference database.

9 . A method as claimed in claim 4 comprising allocating the quantitative image data to a tissue type and thereafter supplying said quantitative image data with said allocation to at least one of said first reference database and said second reference database.

10 . A method as claimed in claim 9 comprising allocating said quantitative image data to said tissue type by selecting a pixel in said quantitative image data and then seeking adjacent pixels having a same quantitative value up to a predetermined deviation, and allocating said pixel and any of said pixels not exceeding said predetermined deviation to said tissue type.

11 . A method as claimed in claim 4 comprising determining a quantitative ratio between at least two physical variables of said predetermined object and displaying a deviation of said quantitative ratio as at least one of a graphic, a color chart, or a color overlay on a medical image corresponding to said quantitative image data.

12 . A method as claimed in claim 1 comprising, in said computer, determining a volume of said predetermined object from said physical variable in the quantitative image data, and comparing said volume with a reference database that comprises a plurality of sets of volume data respectively for different examination objects, and when a predetermined deviation of said volume from one of said volume data sets is exceeded, displaying a designation at said display screen that said deviation has been exceeded.

13 . A method as claimed in claim 1 comprising implementing said segmenting using a statistical model of the examination object that is dependent on at least one statistical variable that is associated with said physical variable.

14 . A method as claimed in claim 1 comprising determining said physical variable by MR fingerprinting.

15 . A method as claimed in claim 1 comprising employing, as said physical variable, a variable selected from the group consisting of T 1 , T 2 , T 2 * and off resonance in magnetic resonance imaging of said quantitative image data.

16 . A method as claimed in claim 1 comprising, after segmenting said quantitative image data, manually allocating healthy and pathological tissue to said quantitative image data via said display screen.

17 . A magnetic resonance apparatus comprising:
a magnetic resonance scanner; a computer configured to operate the magnetic resonance scanner to acquire raw data from an examination object in the magnetic resonance scanner, said examination object exhibiting at least one physical variable; a display monitor in communication with said computer; said computer being configured to automatically determine quantitative image data of the examination object from the raw data and to display said quantitative raw image data at said display monitor as a plurality of pixels with each pixel presenting a value of said at least one physical variable; and said computer being configured to segment the displayed quantitative image data to select at least one segment in the displayed quantitative image data that represents a predetermined image object in said quantitative image data.

Assignees

• Siemens Ag

Inventors

• Bachschmidt Theresa
• Grodzki David
• Meyer Heiko
• Nittka Mathias
• Raithel Esther

Classifications

• G06T7/11

  Region-based segmentation · CPC title

• G06T7/62

  of area, perimeter, diameter or volume · CPC title

• A61B2576/00

  Medical imaging apparatus involving image processing or analysis (A61B1/00009, A61B6/52 and A61B8/52 take precedence) · CPC title

• G06T7/0014

  using an image reference approach · CPC title

• G06T2207/20036

  Morphological image processing · CPC title