TUMOR BOUNDARY RECONSTRUCTION USING HYPERSPECTRAL IMAGING

• US 20200134820A1
• Filed: 10/23/2019
• Published: 04/30/2020
• Est. Priority Date: 10/25/2018
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: HENDRIKS, BERNARDUS HENDRIKUS WILHELMUS, LIU, HONG, SHAN, CAIFENG
• Granted Patent: US 11,257,213 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 2207/10036 Multispectral image; Hyperspectral image
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 2210/12 Bounding box
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/174 involving the use of two or more images

Abstract

In a computer implemented method of determining a boundary of a tumor region or other diseased tissue, hyper- or multispectral image data of a tissue sample including a tumor region or other diseased tissue is taken. The analysis includes a morphological analysis and a spectral analysis of the hyper- or multispectral image data resulting in a morphological tumor boundary and a spectral tumor boundary or a morphological diseased tissue boundary and a spectral diseased tissue boundary. These two boundaries are combined resulting in a combined tumor boundary or combined diseased tissue boundary, wherein an indication of reliability of the combined tumor boundary or combined diseased tissue boundary is given.

Claims

1. Computer implemented method of determining a boundary of a tumor region (120) or other diseased tissue, the method comprising the steps of:

receiving (S1) hyper/multi-spectral image data (140) corresponding to a tissue sample (110) comprising a tumor region (120) or other diseased tissue;

performing (S2) a morphological analysis of the hyper/multi-spectral image data (140) thereby generating a morphological tumor boundary (150) or a morphological diseased tissue boundary corresponding to the tumor region (120) or the other diseased tissue;

performing (S3) a spectral analysis of the hyper/multi-spectral image (140) data thereby generating a spectral tumor boundary (160) or a spectral diseased tissue boundary corresponding to the tumor region (120) or the other diseased tissue;

generating (S4) a combined tumor boundary (170) based on the morphological tumor boundary (150) and the spectral tumor boundary (160) or generating a combined diseased tissue boundary based on the morphological diseased tissue boundary and the spectral diseased tissue boundary.

2. The computer implemented method according to claim 1 wherein the step of performing a morphological analysis of the hyper/multi-spectral image data (140) and/or the step of performing a spectral analysis of the hyper/multi-spectral image data (140) comprises:

processing at least a portion of the hyper/multi-spectral image (140) data with a deep neural network or a machine learning algorithm.

3. The computer implemented method according to claim 2 wherein the step of performing a spectral analysis of the hyper/multi-spectral image data (140) comprises:

classifying a cluster of neighboring pixels as diseased (120) or normal tissue (130).

4. The computer implemented method according to any one of claims 1-3 further comprising the step of:

generating an indication of the reliability (180) of the combined tumor boundary (170) or of the combined diseased tissue boundary.

5. Computer program product comprising instructions which when executed on a processor cause the processor to carry out a method of determining a boundary of a tumor region (120) or other diseased tissue, the method comprising the steps of:

receiving hyper/multi-spectral image data (140) corresponding to a tissue sample (110) comprising a tumor region (120) or other diseased tissue;

performing a morphological analysis of the hyper/multi-spectral image data (140) thereby generating a morphological tumor boundary (150) or a morphological diseased tissue boundary corresponding to the tumor region (120) or the other diseased tissue;

performing a spectral analysis of the hyper/multi-spectral image data (140) thereby generating a spectral tumor boundary (160) or a spectral diseased tissue boundary corresponding to the tumor region (120) or the other diseased tissue;

generating a combined tumor boundary (170) based on the morphological tumor boundary (150) and the spectral tumor boundary (160);

or generating a combined diseased tissue boundary based on the morphological diseased tissue boundary and the spectral diseased tissue boundary.

6. Medical image classification system (100) for determining a boundary of a tumor region (120) or other diseased tissue, the medical image classification system comprising:

a receiving device (190) configured to receive spectral image data (140) corresponding to a tissue sample (110) comprising a tumor region (120) or other diseased tissue;

a morphological analysis device configured to perform a morphological analysis of the spectral image data (140) thereby generating a morphological tumor boundary (150) or a morphological diseased tissue boundary corresponding to the tumor region (120) or the other diseased tissue;

a spectral analysis device configured to perform a spectral analysis of the spectral image data (140) thereby generating a spectral tumor boundary (160) or a spectral diseased tissue boundary corresponding to the tumor region (120) or the other diseased tissue; and

a combination device configured for generating a combined tumor boundary (170) based on the morphological tumor boundary (150) and the spectral tumor boundary (160) or for generating a combined diseased tissue boundary based on the morphological diseased tissue boundary and the spectral diseased tissue boundary.

7. The medical image classification system (100) according to claim 6,

wherein the spectral image data (140) is multispectral image data or hyperspectral image data.

8. The medical image classification system (100) according to any of claims 6 or 7,

wherein the morphological analysis and/or the spectral analysis of the spectral image data comprises the processing of at least a portion of the spectral image data (140) with a deep neural network or a machine learning algorithm or an artificial intelligence module.

9. The medical image classification system according to any of claims 6 to 8,

wherein the spectral analysis of the spectral image data comprises the classifying of a cluster of neighboring pixels as diseased (120) or normal tissue (130).

10. The medical image classification system (100) according to any of claims 6 to 9,

wherein generating the combined tumor boundary (170) comprises generating an indication of reliability (180) of the combined tumor boundary (170) or wherein generating the combined diseased tissue boundary comprises generating an indication of reliability of the combined diseased tissue boundary.

11. The medical image classification system (100) according to claim 10,

wherein the system is configured for dividing the spectral image (140) into subparts, and

wherein the indication of the reliability (180) of the combined tumor boundary (170) is based on a level of similarity of the morphological tumor boundary (150) and the spectral tumor boundary (160) in each subpart of the spectral image (140);

or wherein the indication of the reliability of the combined diseased tissue boundary is based on a level of similarity of the morphological diseased tissue boundary and the spectral diseased tissue boundary in each subpart of the spectral image.

12. The medical image classification system (100) according to any of claim 10 or 11,

wherein the system is configured for presenting the indication of reliability (180) of the combined tumor boundary (170) or the combined diseased tissue boundary in an image on an image display as color-coding.

13. The medical image classification system (100) according to any of claim 10 or 11,

wherein the system is configured for presenting the indication of reliability (180) of the combined tumor boundary (170) or the combined diseased tissue boundary in an image on an image display as a line with variable thickness or structure.

14. The medical image classification system (100) according to claim 13,

wherein the system is configured for coding the indication of reliability (180) of the combined tumor boundary (170) or the combined diseased tissue boundary with a line having a thickness that corresponds to a confidence or an accuracy with which the combined tumor boundary (170) or the combined diseased tissue boundary has been determined.

15. The medical image classification system (100) according to claim 13,

wherein the system is configured for providing a boundary with a highest level of confidence as a solid line and a boundary with a lower level of confidence as a dashed line.