System and method for cardiac imaging

• US 7,450,746 B2
• Filed: 07/26/2006
• Issued: 11/11/2008
• Est. Priority Date: 06/07/2002
• Status: Expired due to Fees

• Current Assignee: Verathon Incorporated (Roper Technologies, Inc.)
• Original Assignee: Verathon Incorporated (Roper Technologies, Inc.)
• Inventors: Dudycha, Stephen, McMorrow, Gerald, Yang, Fuxing, Yuk, Jongtae, Chalana, Vikram, Shankle, Steven J.
• Primary Examiner(s): Kassa; Yosef
• Application Number:
  US11/460,182
• Publication Number:
  US 20080181479A1
• Time in Patent Office: 839 Days
• Field of Search: 382/128, 382/130, 382/131, 382/132
• US Class Current: 382/131
• CPC Class Codes: A61B 5/204 Determining bladder volume
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
• CPC Class Codes: A61B 8/08 Detecting organic movements or changes, e.g. tumours, cysts, swellings A61B8/02 - A61B8/06 take precedence
• CPC Class Codes: A61B 8/0883 for diagnosis of the heart
• CPC Class Codes: A61B 8/4455 Features of the external shape of the probe, e.g. ergonomic aspects
• CPC Class Codes: A61B 8/483 involving the acquisition of a 3D volume of data
• CPC Class Codes: A61B 8/488 involving Doppler signals

Abstract

Ultrasound systems and methods are described to measure changes in cardiac chamber volumes and organ wall areas, thicknesses, volumes and masses between the cardiac chambers using computer readable media employing image processing algorithms applied to 3D data sets acquired at systole and diastole. The systems for cardiac imaging includes an ultrasound transceiver configured to sense the mitral valve of a heart by Doppler ultrasound, an electrocardiograph connected with a patient and synchronized with the transceiver to acquire ultrasound-based 3D data sets during systole and diastole at a transceiver location determined by Doppler ultrasound affected by the mitral valve, and a computer readable medium configurable to process ultrasound imaging information from the 3D data sets communicated from the transceiver.

Claims

1. A method for cardiac imaging, comprising:

creating a database of 3D images having manually segmented regions;

training level-set image processing algorithms to substantially reproduce the shapes of the manually segmented regions using a computer readable medium;

acquiring a non-database 3D image;

segmenting the regions of the non-database image by applying the trained level-set processing algorithms using the computer readable medium, and

determining from the segmented non-database 3D image at least one of;

a volume of any heart chamber, and

a thickness of the wall between any adjoining heart chambers.

2. The method of claim 1, wherein the manually segmented regions of the database of 3D images is determined by deriving information from the data content of the 3D images using a computer readable medium.

3. The method of claim 2, wherein deriving information using the computer readable medium further includes applying image processing algorithms.

4. The method of claim 3, wherein the image processing algorithms include image enhancement, segmentation, combine and clean-up.

5. The method of claim 4, wherein image enhancement includes sub-algorithms heat filter and shock filter, the heat filter being applied first, then followed by the shock filter.

6. The method of claim 5, wherein segmentation includes sub-algorithms intensity-based segmentation and edge-based segmentation, the intensity and edge-based segmentation being applied separately to the results of the shock filter.

7. The method of claim 5, wherein combine includes combining the results of intensity-based and edge-based segmentation sub-algorithms using a Boolean AND operator to produce a segmented image.

8. The method of claim 6, wherein clean-up includes at least one of filling gaps with pixels and removing pixel groups unrelated to at least one region of interest of the image.

9. The method of claim 4, wherein image enhancement includes sub-algorithms radon transfer, heat filter and shock filter, the radon transfer being applied first, followed by the heat filter, and then followed by shock filter.

10. The method of claim 9, wherein segmentation includes sub-algorithms region-based segmentation and gradients-based segmentation, the region and gradient-based segmentation being applied separately to the results of the shock filter.

11. The method of claim 10, wherein the region-based segmentation includes estimate shadow regions, automatic region threshold, and remove shadow regions, the shadow regions being applied to the results of the shock filter, followed by the automatic region threshold, and then followed by the remove shadow regions.

12. The method of claim 10, wherein the gradient-based segmentation includes spatial gradients and hysteresis threshold of gradients, the spatial gradients being applied to the results of the shock filter followed by the hysteresis threshold of gradients.

13. The method of claim 12, wherein spatial gradients further includes kernel density estimate.

14. The method of claim 12, wherein hysteresis threshold further includes non-parametric eigen shape estimation.

15. The method of claim 5, wherein combine includes combining the results of region-based and threshold-based segmentation sub-algorithms using a Boolean AND operator to produce a segmented image.

16. The method of claim 15, wherein cleanup includes at least one of morphological cleanup and snakes-based cleanup, wherein morphological cleanup includes at least one of filling gaps with pixels and removing pixel groups unrelated to at least one region of interest of the image and snakes-based cleanup being applied after morphological cleanup.

17. The method of claim 1, wherein training level-set image processing algorithms includes a training sub-algorithm and a segmentation sub-algorithm.

18. The method of claim 17, wherein the training sub-algorithm includes, in order, acquire apriori training shapes from the database, align the shapes by gradient decent, generate signed distance map, and principle component analysis.

19. The method of claim 17, wherein the segmentation sub-algorithm includes in order minimize shape parameters by gradient descent, update shape image value, determine inside and outside curvature lines from update, and curvature line convergence check.

20. The method of claim 1, wherein acquiring a non-database 3D image includes an ultrasound 3D data set comprising at least one of a plurality of scan planes and a plurality of 3D distributed scanlines comprised of ultrasound echoes collected by an ultrasound transceiver having a speaker.

21. The method of claim 20, wherein acquiring 3D data sets is achieved by an electrocardiograph in signal communication with the transceiver to acquire the 3D data sets at systole and diastole.

22. The method of claim 21, wherein acquiring a non-database 3D image further includes placing an ultrasound transceiver against a subject to transmit radio frequency ultrasound into the subject to detect a mitral valve of a heart from echoic pulses reflected from the heart.

23. The method of claim 22, wherein placing further includes re-positioning the transceiver in response to audible sounds emanating from the speaker that characterizes the centered location as determined by change in audio signals emanating from the speaker in proportion to a Doppler shift experienced by the received echoes.

24. The method of claim 1, wherein segmenting the regions of the non-database image by applying the trained image processing algorithms using the computer readable medium includes level-set sub-algorithms configured to determine the active contours and minimal distance curves of regions of interest of the non-database image using apriori geometric models of the regions of interest contained within the database.

25. The method of claim 22, wherein the level-set sub-algorithms further includes pose parameters in the form of a metric set of Cartesian and angular definitions of pixel locations constituting the 3D data set of the non-database image.

26. The method of claim 22, wherein determining from the segmented non-database 3D images acquired during systole and diastole further includes determining the change in volume of any heart chamber expressed as an ejection fraction.

27. The method of claim 26, wherein determining the change in heart chamber volume between systole and diastole cycles is expressed as a quotient of the difference between heart chamber volumes occurring at systole and diastole cycles and the volume of the chamber at diastole.

28. The method of claim 26, wherein determining the ejection fraction includes the ejection fraction for the left ventricle chamber expressed as a quotient of the difference between left ventricle volumes occurring at systole and diastole and the volume of the left ventricle at diastole.

29. A method for cardiac imaging comprising:

creating a database of 3D images having manually segmented regions;

training level-set image processing algorithms to substantially reproduce the shapes of the manually segmented regions using a computer readable medium;

acquiring non-database 3D images at systole and diastole;

segmenting the regions of the non-database images by applying the trained level-set processing algorithms using the computer readable medium, and

determining from the segmented non-database 3D images at least one of;

a volume of any heart chamber,

a thickness of the wall between any adjoining heart chambers, and

a change in volume in any heart chamber between systole and diastole.

30. A system for cardiac imaging comprising:

a database of 3D images having manually segmented regions;

an ultrasound transceiver configured to deliver ultrasound pulses into and acquire ultrasound echoes from a subject as 3D image data sets;

an electrocardiograph to determine the timing to acquire the 3D data sets; and

a computer readable medium configured to train level-set image processing algorithms to substantially reproduce the shapes of the manually segmented regions and to segment regions of interest of the 3D data sets using the trained algorithms,

wherein at least one cardiac metric from the 3D data sets is determined from the segmented regions of interest.

31. The system of claim 30, wherein the ultrasound transceiver includes a speaker configurable to sense a Doppler shift of the ultrasound echoes affected by the movement of the mitral valve of the heart.

32. The system of claim 31, wherein the location to acquire 3D data sets is determined by the change in audible signals from the speaker in proportion to the Doppler shift.

33. The system of claim 32, wherein the 3D data sets are acquired during systole and diastole.

34. The system of claim 33, wherein the at least one cardiac metric further includes the change in volume of any heart chamber, the change in thickness of the wall between any adjoining heart chambers, and the change in volume of the wall between any adjoining heart chambers.

35. The system of claim 30, wherein the computer readable medium configured to train level-set image processing algorithms to substantially reproduce the shapes of the manually segmented regions includes a training sub-algorithm and a segmentation sub-algorithm.

36. The system of claim 35, wherein the training sub-algorithm includes, in order, acquire apriori training shapes from the database, align the shapes by gradient decent, generate signed distance map, and principle component analysis.

37. The system of claim 35, wherein the segmentation sub-algorithm includes in order minimize shape parameters by gradient descent, update shape image value, determine inside and outside curvature lines from update, and curvature line convergence check.

38. The system of claim 37, wherein the segmentation sub-algorithm includes in order minimize shape parameters by gradient descent, update shape image value, determine inside and outside curvature lines from update, and curvature line convergence check.

39. The system of claim 30, wherein the computer readable medium configured to segment regions of interest of the 3D data sets using the trained algorithms using includes level set algorithms that determine active contours and minimal distance curves of regions of interest of the 3D data sets using apriori geometric models of the regions of interest contained within the database.

40. The system of claim 39, wherein the at least one cardiac metric includes determining the ejection fraction of at least one heart chamber defined as the quotient of the change in the at least one heart chamber volume between systole and diastole cycles and the volume of the same at least one heart chamber at diastole.

41. The system of claim 40, wherein the at least one chamber includes a left ventricle, a right ventricle, a left atrium, and a right atrium.

42. A system for cardiac imaging comprising:

an ultrasound transceiver configured to determine the location to acquire ultrasound echo 3D data sets to image the left ventricle of a heart;

an electrocardiograph to determine the timing to acquire the 3D data sets at a systole and diastole; and

a computer readable medium configured to train level-set image processing algorithms to substantially reproduce the shapes of the manually segmented regions and to segment regions of interest of the 3D data sets using the trained algorithms to determine between systole and diastole at least one of;

the thickness of a wall between the left ventricle and at least one of the adjoining heart chambers,

the volume of the left ventricle, and

the ejection fraction of the left ventricle.