Determining regions of hyperdense lung tissue in an image of a lung

Patent metadata

Field	Value
Publication number	US-11348229-B2
Application number	US-201816644037-A
Country	US
Kind code	B2
Filing date	Sep 4, 2018
Priority date	Sep 5, 2017
Publication date	May 31, 2022
Grant date	May 31, 2022

Abstract

There is provided a computer-implemented method and system (100) for determining regions of hyperdense lung parenchyma in an image of a lung. The system (100) comprises a memory (106) comprising instruction data representing a set of instructions and a processor (102) configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor (102), cause the processor (102) to locate a vessel in the image, determine a density of lung parenchyma in a region of the image that neighbours the located vessel, and determine whether the region of the image comprises hyperdense lung parenchyma based on the determined density, hyperdense lung parenchyma having a density greater than −800 HU.

First claim

The invention claimed is:

1. A system configured for determining regions of hyperdense lung parenchyma in an image of a lung, the system comprising:
a memory comprising instruction data representing a set of instructions;
a processor configured to communicate with the memory and to execute the set of instructions, wherein the set of instructions, when executed by the processor, cause the processor to:
locate a vessel in the image by locating a vessel stump and iteratively locating further portions of the vessel from the location of the vessel stump based on one or more of direction of the further portions of the vessel or distance from the vessel stump of the further portions of the vessel;
identify a region of lung tissue to evaluate for hyperdense lung parenchyma by selecting only regions that neighbor the located vessel, wherein the neighboring regions comprise connected image components that are adjacent to the located vessel;
determine a density of lung parenchyma in the identified region; and
determine whether the identified region comprises hyperdense lung parenchyma based on the determined density, hyperdense lung parenchyma having a density greater than −800 Hounsfield units (HU).

2. The system as claimed in claim 1 , wherein the image comprises a plurality of image components; and
wherein the determined density of lung parenchyma in the region is a distribution of density values of the image components in the region.

3. The system as claimed in claim 2 , wherein the distribution of density values comprises a histogram of density values of the image components in the region.

4. The system as claimed in claim 1 wherein, if the region is determined to comprise hyperdense lung parenchyma, the set of instructions, when executed by the processor, further cause the processor to:
identify a medical condition associated with the hyperdense lung parenchyma using a distribution of density values in the region of the image that neighbors the located vessel.

5. The system as claimed in claim 1 , wherein the set of instructions, when executed by the processor, further cause the processor to:
determine a location of a bone in the image; and
determine that hyperdense lung parenchyma is absent in the region if the region overlaps the determined location of the bone.

6. The system as claimed in claim 1 , wherein the set of instructions, when executed by the processor, further cause the processor to:
determine a location of a hyperdense region in the image; and
determine that the hyperdense region comprises pleural effusion if the location of the hyperdense region lies outside the region of the image that neighbors the located vessel.

7. The system as claimed in claim 1 , wherein a region is determined to comprise hyperdense lung parenchyma where the density of the lung parenchyma in the region is greater than an average density of aerated lung parenchyma.

8. The system as claimed in claim 7 , wherein the average density of aerated lung parenchyma is approximately −800 Hounsfield units (HU).

9. The system as claimed in claim 1 , wherein causing the processor to locate a vessel in the image comprises causing the processor to:
segment the image to locate one or more portions of the heart;
identify, from the segmentation, the vessel stump, wherein the vessel stump corresponds to a portion of the vessel that leaves the heart; and
locate the vessel in the image using the stump as a starting point.

10. The system as claimed in claim 9 , wherein the image comprises a plurality of image components and wherein causing the processor to locate a vessel in the image comprises causing the processor to:
for each image component in a region surrounding the vessel stump, determine a measure that is indicative of the likelihood that the image component comprises part of the vessel; and
locate a further portion of the vessel from the location of the vessel stump and the determined measures.

11. The system as claimed in claim 10 , wherein causing the processor to locate the vessel in the image further comprises causing the processor to iteratively, for each further portion of the vessel that is located:
for each image component in a region surrounding the further portion of the vessel, determine a measure that is indicative of the likelihood that the image component comprises part of the vessel; and
locate another further portion of the vessel from the determined measures.

12. The system as claimed in claim 1 , wherein the processor is caused to locate a vessel in the image with a diameter that is less than a predefined threshold diameter.

13. A computer-implemented method for determining regions of hyperdense lung parenchyma in an image of a lung, the method comprising:
locating a vessel in the image by locating a vessel stump and iteratively locating further portions of the vessel from the location of the vessel stump based on one or more of direction of the further portions of the vessel or distance from the vessel stump of the further portions of the vessel;
identifying a region of lung tissue to evaluate for hyperdense lung parenchyma by selecting only regions that neighbor the located vessel, wherein the neighboring regions comprise connected image components that are adjacent to the located vessel;
determining a density of lung parenchyma in the identified region of the image; and
determining whether the identified region comprises hyperdense lung parenchyma based on the determined density, hyperdense lung parenchyma having a density greater than −800 Hounsfield units (HU).

14. A non-transitory computer readable medium having computer readable code embodied therein, the computer readable code being configured such that, on execution by a suitable computer or processor, causes the computer or processor to:
locate a vessel in the image by locating a vessel stump and iteratively locating further portions of the vessel from the location of the vessel stump based on one or more of direction of the further portions of the vessel or distance from the vessel stump of the further portions of the vessel;
identify a region of lung tissue to evaluate for hyperdense lung parenchyma by selecting only regions that neighbor the located vessel, wherein the neighboring regions comprise connected image components that are adjacent to the located vessel;
determine a density of lung parenchyma in the identified region of the image; and
determine whether the identified region comprises hyperdense lung parenchyma based on the determined density, hyperdense lung parenchyma having a density greater than −800 Hounsfield units (HU).

Assignees

• Koninklijke Philips Nv

Inventors

• Wiemker Rafael
• Saalbach Axel
• Von Berg Jens
• Brosch Tom
• Harder Tim Philipp
• Wenzel Fabian
• Hall Christopher Stephen

Classifications

• G06V20/00 Primary

  Scenes; Scene-specific elements (control of digital cameras H04N23/60) · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• G06V10/507

  Summing image-intensity values; Histogram projection analysis · CPC title

• G06T2207/30008

  Bone · CPC title

• G06T2207/30048

  Heart; Cardiac · CPC title