Method and apparatus for detecting interventricular dyssynchrony

• US 8,131,043 B2
• Filed: 09/18/2006
• Issued: 03/06/2012
• Est. Priority Date: 09/16/2005
• Status: Expired due to Fees

• Current Assignee: Ohio State University
• Original Assignee: Ohio State University
• Inventors: Binkley, Philip F., Raman, Subha V.
• Primary Examiner(s): Taningco, Alexander H
• Application Number:
  US12/067,092
• Publication Number:
  US 20080253638A1
• Time in Patent Office: 1,996 Days
• Field of Search: 382/128, 382/131, 378/8
• US Class Current: 382/131
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: A61B 5/1075 for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer A61B8/0858 takes precedence
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 6/503 for diagnosis of the heart
• CPC Class Codes: A61B 8/0883 for diagnosis of the heart
• CPC Class Codes: G06T 2207/10076 4D tomography; Time-sequential 3D tomography
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/30048 Heart; Cardiac
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/62 of area, perimeter, diameter or volume
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Several embodiments of a method and apparatus for detecting interventricular dyssynchrony are provided. In one embodiment, the apparatus includes an image segment processor and an image classifier. The image segment processor may segment left and right ventricles in source images of a subject'"'"'s heart to form left and right ventricle segments in segmented images. Each source image may include cross sections of the right and left ventricles in common spatial relation along the short axis plane and in common temporal relation in reference to a cardiac cycle. The source images and corresponding segmented images may be temporally-spaced in relation to the cardiac cycle. The image classifier may determine first and second cross-sectional areas associated with the left and right ventricle segments for each segmented image, compare the first and second cross-sectional areas, and classify the subject'"'"'s heart in an interventricular dyssynchronous class or a non-dyssynchronous class.

Claims

1. An apparatus for detecting interventricular dyssynchrony, including:

an image segment processor segmenting a left ventricle and a right ventricle in each of a plurality of source images of a heart to form left and right ventricle segments in a corresponding plurality of segmented images, each source image including at least a portion of a cross section of the heart along a short axis plane, each source image including cross sections of the right and left ventricles of the heart in common spatial relation along the short axis plane and in common temporal relation in reference to a cardiac cycle associated with the heart, the plurality of source images and corresponding plurality of segmented images being temporally-spaced in relation to the cardiac cycle; and

an image classifier, in operative communication with the image segment processor, determining a first cross-sectional area associated with the left ventricle segment and a second cross-sectional area associated with the right ventricle segment for each segmented image, comparing, at least in part based on use of a linear regression model, the first and second cross-sectional areas for each segmented image, and classifying the heart in an interventricular dyssynchronous class or a non-dyssynchronous class, based at least in part on the comparing, by classifying the heart in the non-dyssynchronous class if the linear regression model attains statistical significance and indicates a linear relation between the first and second cross-sectional areas over the cardiac cycle and by classifying the heart in the interventricular dyssynchronous class if the linear regression model indicates there is not a linear relation between the first and second cross-sectional areas over the cardiac cycle.

2. The apparatus of claim 1, further including:

an image sequence processor, in operative communication with the image segment processor, generating the plurality of source images of the heart from image data resulting from detecting characteristics of the heart.

3. The apparatus of claim 2, further including:

a medical imaging device, in operative communication with the image sequence processor, generating the image data by detecting the characteristics of the heart.

4. The apparatus of claim 3 wherein the medical imaging device includes at least one of a medical radiography device, a medical tomography device, a medical sonography device, and a nuclear medicine device.

5. The apparatus of claim 4 wherein the medical imaging device includes a magnetic resonance imaging device, an echocardiography device, a multi-detector computed tomography device, a single photon emission computed tomography device, or a positron emission tomography device.

6. The apparatus of claim 1, further including:

a network interface, in operative communication with the image segment processor, adapted to receive the plurality of source images from a medical imaging system via a communication network.

7. A method of detecting interventricular dyssynchrony, including:

a) receiving a plurality of source images of a heart, each source image including at least a portion of a cross section of the heart along a short axis plane, each source image including cross sections of right and left ventricles of the heart in common spatial relation along the short axis plane and in common temporal relation in reference to a cardiac cycle associated with the heart, the plurality of source images being temporally-spaced in relation to the cardiac cycle;

b) segmenting the left ventricle and the right ventricle in each source image to form left and right ventricle segments in a corresponding plurality of segmented images;

c) determining a first cross-sectional area associated with the left ventricle segment and a second cross-sectional area associated with the right ventricle segment for each segmented image;

d) comparing, at least in part based on use of a linear regression model, the first and second cross-sectional areas for each segmented image;

e) classifying the heart in an interventricular dyssynchronous class or a non-dyssynchronous class based at least in part on the comparing, the classifying including;

f) classifying the heart in the non-dyssynchronous class if the linear regression model attains statistical significance and indicates a linear relation between the first and second cross-sectional areas over the cardiac cycle; and

g) classifying the heart in the interventricular dyssynchronous class if the linear regression model indicates there is not a linear relation between the first and second cross-sectional areas over the cardiac cycle.

8. The method of claim 7 wherein the plurality of source images are received from a medical imaging system via a communication network.

9. The method of claim 7 wherein the spatial relation of the plurality of source images in the short axis plane is associated with a mid-cavity view of the heart.

10. The method of claim 7, further including:

h) identifying a first boundary defining the left ventricle segment and a second boundary defining the right ventricle segment in each source image.

11. The method of claim 10, further including:

i) detecting edges in each source image;

j) in each source image, selecting one or more detected edges defining at least a portion of the left ventricle segment; and

k) in each source image, selecting one or more detected edges defining at least a portion of the right ventricle segment.

12. The method of claim 11, further including:

l) in each source image, connecting the detected edges defining at least a portion of the left ventricle segment to form the first boundary; and

m) in each source image, connecting the detected edges defining at least a portion of the right ventricle segment to form the second boundary.

13. The method of claim 10 wherein the first boundary is associated with an outer wall of the left ventricle and the second boundary is associated with an outer wall of the right ventricle.

14. The method of claim 7 wherein the first cross-sectional area is determined using a first algorithm and the second cross-sectional area is determined using a second algorithm.

15. The method of claim 7, further including:

h) diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying; and

i) planning treatment of the heart to include biventricular pacing through implantation of a cardiac resynchronization therapy device.

16. The method of claim 15, further including:

j) planning for the cardiac resynchronization therapy device to operate in a sequential biventricular pacing mode with an initial interventricular delay value based at least in part on results of the linear regression model.

17. The method of claim 7, further including:

h) classifying the heart in the interventricular dyssynchronous class if results of the linear regression model fail to attain statistical significance.

18. The method of claim 7, further including:

h) pre-diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying.

19. The method of claim 7, further including:

h) diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying; and

i) planning treatment of the heart to include implantation of a cardiac resynchronization therapy device.

20. The method of claim 7 wherein the segmenting in b) is based at least in part on observation of the plurality of source images by a qualified healthcare professional.

21. The method of claim 7 wherein at least one of the determining in c), comparing in d), and classifying in at least one of e), f), and g) is based at least in part on observation of the plurality of segmented images by a qualified healthcare professional.

22. The method of claim 21, further including:

h) pre-diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying.

23. The method of claim 21, further including:

h) diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying;

i) planning treatment of the heart to include implantation of a cardiac resynchronization therapy device.

24. A method of detecting interventricular dyssynchrony, including:

a) receiving one or more source images of a heart, each source image including at least a portion of a cross section of the heart along a short axis plane, the one or more source images including cross sections of right and left ventricles of the heart in common spatial relation along the short axis plane and in common temporal relation to a cardiac cycle associated with the heart;

b) segmenting the left ventricle and the right ventricle from the one or more source images to form left and right ventricle segments in one or more segmented images;

c) determining a first cross-sectional area associated with the left ventricle segment and a second cross-sectional area associated with the right ventricle segment from the one or more segmented images;

d) comparing, at least in part based on use of a linear regression model, the first and second cross-sectional areas; and

e) classifying the heart in an interventricular dyssynchronous class or a non-dyssynchronous class based at least in part on the comparing, the classifying including;

f) classifying the heart in the non-dyssynchronous class if the linear regression model attains statistical significance and indicates a linear relation between the first and second cross-sectional areas over the cardiac cycle; and

g) classifying the heart in the interventricular dyssynchronous class if the linear regression model indicates there is not a linear relation between the first and second cross-sectional areas over the cardiac cycle.

25. The method of claim 24, further including:

h) pre-diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying.

26. The method of claim 24, further including:

h) diagnosing the heart as exhibiting interventricular dyssynchrony based at least in part on the classifying;

i) planning treatment of the heart to include implantation of a cardiac resynchronization therapy device.

27. The method of claim 24 wherein at least one of the determining in c), comparing in d), or classifying in at least one of e), f), and g) is based at least in part on observation of the segmented images by a qualified healthcare professional.

28. An apparatus for detecting interventricular dyssynchrony, including:

an image segment processor segmenting a left ventricle and a right ventricle from one or more source images of a heart to form left and right ventricle segments in one or more segmented images, each source image including at least a portion of a cross section of the heart along a short axis plane, the one or more source images, at least in combination, including cross sections of right and left ventricles of the heart in common spatial relation along the short axis plane and in common temporal relation in reference to a cardiac cycle associated with the heart; and

an image classifier, in operative communication with the image segment processor, determining a first cross-sectional area associated with the left ventricle segment and a second cross-sectional area associated with the right ventricle segment, comparing, at least in part based on use of a linear regression model, the first and second cross-sectional areas, and classifying the heart in an interventricular dyssynchronous class or a non-dyssynchronous class based at least in part on the comparing, by classifying the heart in the non-dyssynchronous class if the linear regression model attains statistical significance and indicates a linear relation between the first and second cross-sectional areas over the cardiac cycle and by classifying the heart in the interventricular dyssynchronous class if the linear regression model indicates there is not a linear relation between the first and second cross-sectional areas over the cardiac cycle.

29. The apparatus of claim 28, further including:

an image sequence processor, in operative communication with the image segment processor, generating the one or more source images of the heart from image data resulting from detecting characteristics of the heart.

30. The apparatus of claim 29, further including:

a medical imaging device, in operative communication with the image sequence processor, generating the image data by detecting the characteristics of the heart.

31. The apparatus of claim 28, further including:

a network interface, in operative communication with the image segment processor, adapted to receive the one or more source images from a medical imaging system via a communication network.