System and method for staging regional lymph nodes using quantitative analysis of endoscopic ultrasound images

• US 6,154,560 A
• Filed: 08/30/1996
• Issued: 11/28/2000
• Est. Priority Date: 08/30/1996
• Status: Expired due to Fees

• Current Assignee: Cleveland Clinic Foundation
• Original Assignee: Cleveland Clinic Foundation
• Inventors: Cothren, Robert M., Zuccaro, Gregory Jr., Brandao, Alexandre Purper
• Primary Examiner(s): Chang, Jon
• Assistant Examiner(s): Patel, Jayanti K
• Application Number:
  US08/706,421
• Time in Patent Office: 1,551 Days
• Field of Search: 128/915, 128/916, 128/660.01, 128/662.06, 128/922, 382/108, 382/128-134, 382/203, 382/204, 395/118, 395/129, 395/130, 600/437, 600/443, 600/463, 250/461.2
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/10068 Endoscopic image
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 2207/30101 Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A system and method for automatically staging regional lymph nodes under the TNM classification system using textural analyses of endoscopic ultrasound images. N stage for a lymph node is determined by analyzing the organization of the texture of a region corresponding to the node on a digitized ultrasound image. Specifically, Haralick'"'"'s difference entropy and correlation textural features are calculated and compared to predetermined reference values to predict whether the lymph node is benign or malignant. Also, the cross-sectional area of the node is determined and used to make an N staging determination. An application of the present invention is staging of esophageal lymph nodes in the diagnosis and treatment of esophageal carcinoma using endoscopic ultrasound images.

Claims

1. A method for analyzing an ultrasound image of a lymph node to determine the status of the lymph node as being at a particular stage, comprising the steps of:

(a) converting the ultrasound image into a digital ultrasound image;

(b) determining a region of the digital ultrasound image corresponding to the lymph node;

(c) determining at least one arithmetic value related to the texture of at least a portion of the region of the digital ultrasound image corresponding to the lymph node;

(d) comparing the at least one arithmetic value to at least one predetermined value; and

(e) determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value.

2. A method for analyzing an ultrasound image of a lymph node according to claim 1 wherein said step of determining at least one arithmetic value comprises the step of determining at least one arithmetic value related to the organization of the texture of at least a portion of the region of the digital ultrasound image corresponding to the lymph node.

3. A method for analyzing an ultrasound image of a lymph node according to claim 1 wherein said step of determining at least one arithmetic value comprises the step of determining at least one arithmetic value related to the uniformity of the texture of at least a portion of the region of the digital ultrasound image corresponding to the lymph node.

4. A method for analyzing an ultrasound image of a lymph node according to claim 1 wherein said step of determining at least one arithmetic value comprises the step of determining a difference entropy value related to a difference entropy textural feature for at least a portion of the region of the digital ultrasound image corresponding to the lymph node.

5. A method for analyzing an ultrasound image of a lymph node according to claim 1 wherein said step of determining at least one arithmetic value comprises the step of determining an entropy value related to an entropy textural feature for at least a portion of the region of the digital ultrasound image corresponding to the lymph node.

6. A method for analyzing an ultrasound image of a lymph node according to claim 1 wherein said step of determining at least one arithmetic value comprises the step of determining a correlation value related to a correlation textural feature for at least a portion of the region of the digital ultrasound image corresponding to the lymph node.

7. A method for analyzing an ultrasound image of a lymph node according to claim 1 wherein said step of determining at least one arithmetic value comprises the step of determining a difference entropy value related to a difference entropy textural feature and a correlation value related to a correlation textural feature for at least a portion of the region of the digital ultrasound image corresponding to the lymph node.

8. A method for analyzing an ultrasound image of a lymph node according to claim 1 further comprising the steps of:

(a) determining an arithmetic size value related to the size of the region of the digital ultrasound image corresponding to the lymph node;

(b) comparing the arithmetic size value to at least one predetermined size value; and

(c) determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value and the relationship between the arithmetic size value and the at least one redetermined size value.

9. A method for analyzing an ultrasound image of a lymph node according to claim 8 wherein said step of determining an arithmetic size value comprises the step of determining an arithmetic area value related to the area of the region of the digital ultrasound image corresponding to the lymph node.

10. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node to determine the status of the lymph node as being at a particular stage, comprising the steps of:

(a) gathering position-correlated signals of a predetermined parameter of tissue including a lymph node;

(b) converting the position-correlated signals of a predetermined parameter of tissue including a lymph node into a matrix of digital parameter data;

(c) determining a region of the matrix corresponding to the lymph node;

(d) determining at least one arithmetic value related to the texture of at least a portion of the region of the matrix corresponding to the lymph node;

(e) comparing the at least one arithmetic value to at least one predetermined value; and

(f) determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value.

11. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said matrix is a two-dimensional matrix representing position-correlated signals of a predetermined parameter for a slice of tissue.

12. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said step of determining at least one arithmetic value comprises the step of determining at least one arithmetic value related to the organization of the texture of at least a portion of the region of the matrix corresponding to the lymph node.

13. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said step of determining at least one arithmetic value comprises the step of determining at least one arithmetic value related to the uniformity of the texture of at least a portion of the region of the matrix corresponding to the lymph node.

14. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said step of determining at least one arithmetic value comprises the step of determining a difference entropy value related to a difference entropy textural feature for at least a portion of the region of the matrix corresponding to the lymph node.

15. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said step of determining at least one arithmetic value comprises the step of determining an entropy value related to an entropy textural feature for at least a portion of the region of the matrix corresponding to the lymph node.

16. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said step of determining at least one arithmetic value comprises the step of determining a correlation value related to a correlation textural feature for at least a portion of the region of the matrix corresponding to the lymph node.

17. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 wherein said step of determining at least one arithmetic value comprises the step of determining a difference entropy value related to a difference entropy textural feature and a correlation value related to a correlation textural feature for at least a portion of the region of the matrix corresponding to the lymph node.

18. A method for analyzing position-correlated signals of a predetermined parameter of a lymph node according to claim 10 further comprising the steps of:

(a) determining an arithmetic size value related to the size of the region of the matrix corresponding to the lymph node;

(b) comparing the arithmetic size value to at least one predetermined size value; and

(c) determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value and the relationship between the arithmetic size value and the at least one predetermined size value.

19. A method for analyzing an ultrasound image of a lymph node according to claim 18 wherein said step of determining an arithmetic size value comprises the step of determining an arithmetic area value related to the area of the matrix corresponding to the lymph node.

20. Apparatus for analyzing an ultrasound image of a lymph node to determine the status of the lymph node as being at a particular stage, comprising:

(a) means for converting the ultrasound image into a digital ultrasound image;

(b) means for determining a region of the digital ultrasound image corresponding to the lymph node;

(c) means for determining at least one arithmetic value related to the organization of the texture of at least a portion of the region of the digital ultrasound image corresponding to the lymph node;

(d) means for comparing the at least one arithmetic value to at least one predetermined value; and

(e) means for determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value.

21. Apparatus for analyzing an ultrasound image of a lymph node according to claim 20 further comprising:

(a) means for determining an arithmetic size value related to the size of the region of the digital ultrasound image corresponding to the lymph node;

(b) means for comparing the arithmetic size value to at least one predetermined size value; and

(c) means for determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value and the relationship between the arithmetic size value and the at least one predetermined size value.

22. A node classifier for determining whether a lymph node is at a particular stage, said node classifier to be interfaced to an ultrasound unit outputting an electrical signal corresponding to an ultrasound image of a slice of tissue including the lymph node, said node classifier comprising:

(a) a processing unit;

(b) a signal interface in circuit communication with said processing unit, said signal interface receiving the electrical signal from the ultrasound unit, generating a received ultrasound image, and transmitting the received ultrasound image to said processing unit;

(c) a texture analyzer in circuit communication with said processing unit, said texture analyzer performing a textural analysis by analyzing the organization of the texture of at least one portion of the received ultrasound image corresponding to the lymph node and determining from the textural analysis whether the lymph node is at the particular stage.

23. A node classifier according to claim 22 wherein said texture analyzer comprises code executing on said processing unit.

24. A node classifier according to claim 22 wherein said texture analyzer comprises circuitry in circuit communication with said processing unit.

25. A node classifier according to claim 22 further comprising a region outliner in circuit communication with said processing unit, said region outliner defining a node region of the received ultrasound image corresponding to the lymph node;

and wherein said texture analyzer analyzes the organization of the texture of the node region.

26. A node classifier according to claim 22 wherein said texture analyzer analyzes a difference entropy textural feature and a correlation textural feature of said at least one portion of the digital ultrasound image corresponding to the lymph node.

27. A node classifier according to claim 25 wherein said texture analyzer analyzes a difference entropy textural feature and a correlation textural feature of the node region.

28. A computer-readable computer program product including a computer readable medium having computer program logic recorded thereon for enabling a processor in a computer system to determine whether a lymph node is at a particular stage from a digital ultrasound image corresponding to the lymph node, said computer-readable computer program product comprising:

(a) means for determining a region of the digital ultrasound image corresponding to the lymph node;

(b) means for determining at least one arithmetic value related to the organization of the texture of at least a portion of the region of the digital ultrasound image corresponding to the lymph node;

(c) means for comparing the at least one arithmetic value to at least one predetermined value; and

(d) means for determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value.

29. A said computer-readable computer program product according to claim 28 further comprising:

(a) means for determining an arithmetic size value related to the size of the region of the digital ultrasound image corresponding to the lymph node;

(b) means for comparing the arithmetic size value to at least one predetermined size value; and

(c) means for determining whether the lymph node is at the particular stage based on the relationship between the at least one arithmetic value and the at least one predetermined value and the relationship between the arithmetic size value and the at least one predetermined size value.

30. A method of staging lymph nodes comprising the steps of:

(a) acquiring and digitizing an endoscopic ultrasound image of at least one lymph node;

(b) defining a portion of the image that corresponds to the at least one lymph node;

(c) analyzing the textural features of the defined portion of the image that corresponds to the at least one lymph node, wherein this step comprises the steps of;

(1) segmenting the defined portion of the image corresponding to the at least one lymph node into a plurality of pixel matrices;

(2) for each pixel matrix, determining a difference entropy textural feature, a correlation textural feature, and a size value for the lymph node;

(3) comparing the difference entropy textural feature, the correlation textural feature, and the size value for the lymph node to a predetermined reference standard; and

(4) determining the stage of the lymph node from the comparison.