Lung measurements

Patent metadata

Field	Value
Publication number	US-9805467-B2
Application number	US-201414898371-A
Country	US
Kind code	B2
Filing date	Jun 19, 2014
Priority date	Jun 19, 2013
Publication date	Oct 31, 2017
Grant date	Oct 31, 2017

Abstract

The present invention relates to lung measurement. In order to provide enhanced information about a patient that facilitates further assessment steps, 2D X-ray image data of a patient's chest is provided, and the image data is segmented to identify lung structures to provide segmented image data separated from un-segmented areas. Further, spatial lung volume information is extracted from the image data using the segmented image data derived from the image data. Still further, lungs symmetry information is determined using the extracted spatial lung volume information. Finally, the lungs symmetry information is provided to a user. For example, a 2D X-ray image data of a patient's chest is provided ( 84 ) and a lungs mask image is formed ( 86 ) after the step of segmenting the input image data. Then, the lungs mask image is used to define areas, within which a predetermined adaptation is applied ( 88 ) to the original 2D X-ray image data producing a thorax mask image. Next, left and right images are provided ( 90 ) showing the left and the right spatial lungs volume information of the regions defined originally by the lungs mask image. Finally, based on the spatial lungs volume information, lungs symmetry information or total lung volumes may be calculated and provided ( 92 ).

First claim

The invention claimed is:

1. A method for providing lungs symmetry information, comprising the following steps:
a) providing 2D X-ray image data of a patient's chest;
b) segmenting the image data to provide segmented image data separated from un-segmented areas, the segmented image data comprising a lungs mask image having first and second selected regions corresponding to lung structures;
c) extracting spatial lung volume information from the image data using the segmented image data derived from the image data;
d) determining lungs symmetry information using the extracted spatial lung volume information; and
e) providing the lungs symmetry information to a user,
wherein the extracting step c) further comprises:
c1) generating a thorax mask image wherein, for the first and second selected regions, the image data is replaced by applying a pixel value representing soft tissue;
c2) subtracting the thorax mask image from the 2D X-ray image data to achieve a lungs-only image comprising the spatial lung volume information, and
wherein the determining step d) further comprises:
d1) summing the spatial lung volume information of the first and second selected regions.

2. Method according to claim 1 , wherein in step a) the 2D X-ray image data is provided as image data acquired in:
a1) a posterior-anterior or an anterior-posterior direction; or
a2) in an oblique direction provided in an angular alignment to the posterior-anterior or anterior-posterior direction, wherein the image data comprises separate image regions of the right and the left lung respectively.

3. Method according to claim 1 , wherein the step b) further comprises:
b i ) identifying left and right lung of the lungs; or
b ii ) identifying a lateral boundary of the lungs enclosing the left and right lungs; and
wherein the extraction in step c) is provided based on:

c i ) the left and the right lung; or
c ii ) the lateral boundary of the lungs.

4. Method according to claim 1 , wherein
step d) further comprises:

d2) applying a conversion factor to the summed spatial lungs volume information to obtain left and right lung volume representations;
d3) calculating a total representation of the left and right lung volume representations; and
d4) generating the lungs symmetry information using the left and right lung volume representations and the total representation.

5. Method according to claim 1 , wherein:
step c1) further comprises the step of interpolating over the image data corresponding to the first and second selected regions of the lungs mask image to remove lungs information, producing a first area (A right ) and a second area (A left ) of the thorax, wherein the first and second areas are located within the thorax mask image; and
step c2) further comprises the step of subtracting the thorax mask image from the image data to produce first spatial lung volume data (L right ) and second spatial lung volume data (L left ).

6. Method according to claim 1 , wherein the conversion factor is a scale using an inverse of a half value layer.

7. A lungs measurement device, comprising:
an interface unit;
a processing unit; and
an output unit;

wherein the interface unit is configured to provide 2D X-ray image data of a patient's chest;
wherein the processing unit is configured to segment the image data to provide segmented image data separated from un-segmented areas, the segmented image data comprising a lungs mask image having first and second selected regions corresponding to lung structures; and to extract spatial lungs volume information from the image data using the segmented image data derived from the image data; and to determine lungs symmetry information using the extracted spatial lungs volume information,
wherein the processing unit is further configured to generate a thorax mask image wherein, for the first and second selected regions, the image data is removed by applying a pixel value representing soft tissue, and to subtract the thorax mask image from the 2D X-ray image data to achieve a lungs-only image comprising the spatial lung volume information, and to sum the spatial lung volume information of the first and second selected regions,
wherein the output unit is configured to provide the lungs symmetry information to a user.

8. Device according to claim 7 , wherein the processing unit is further configured to apply a conversion factor to the summed spatial lungs volume information to obtain left and right lung volume representations; and to calculate a total representation of the left and right lung volume representations; and to generate the lungs symmetry information by the use of the left and right lung volume representations and the total representation.

9. Device according to claim 7 , wherein the processing unit is further configured to interpolate over the image data corresponding to the first and second selected regions to remove lungs information, to produce a first area (A right ) and a second area (A left ) of the thorax, wherein the first and second areas are located within the thorax mask image; and to subtract the thorax mask image from the image data to produce first spatial lung volume data (L right ) and second spatial lung volume data (L left ).

10. An X-ray imaging arrangement, comprising:
an X-ray image acquisition device with an X-ray source and an X-ray detector; and
a lungs measurement device according to claim 7 ;

wherein the X-ray image acquisition device is configured to acquire the image data of the chest of a patient and to provide the image data to an interface of the lungs measurement device.

11. A non-transitory computer-readable storage medium storing a computer program comprising instructions for controlling an X-ray imaging arrangement the X-ray imaging arrangement comprising:
an X-ray image acquisition device with an X-ray source and an X-ray detector; and
a lungs measurement device according to claim 7 ;

wherein the X-ray image acquisition device is configured to acquire the image data of the chest of a patient and to provide the image data to an interface of the lungs measurement device;
which, when being executed by the processing unit, is adapted to perform the method steps of

a) providing 2D X-ray image data of a patient's chest;
b) segmenting the image data to provide segmented image data separated from un-segmented areas, the segmented image data comprising a lungs mask image having first and second selected regions corresponding to lung structures;
c) extracting spatial lung volume information from the image data using the segmented image data derived from the image data;
d) determining lungs symmetry information using the extracted spatial lung volume information; and
e) providing the lungs symmetry information to a user,
wherein the extracting step c) further comprises:
c1) generating a thorax mask image wherein, for the first and second selected regions, the image data is replaced by applying a pixel value representing soft tissue;
c2) subtracting the thorax mask image from the 2D X-ray image data to achieve a lungs only image comprising the spatial lung volume information, and
wherein the determining step d) further comprises:
d1) summing the spatial lung volume information of the first and second selected regions.

Assignees

• Koninklijke Philips Nv

Inventors

• Maack Hanns-Ingo
• Goossen Andre

Classifications

• G06T2207/30061

  Lung · CPC title

• G06T2207/10116

  X-ray image · CPC title

• G06T2207/20224

  Image subtraction · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• G06K9/46

  Physics · mapped topic