Systems and methods for providing automated regional myocardial assessment for cardiac imaging

• US 20050059876A1
• Filed: 06/25/2004
• Published: 03/17/2005
• Est. Priority Date: 06/25/2003
• Status: Active Grant

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Krishnan, Sriram, Rao, R. Bharat, Bennett, Richard M.
• Granted Patent: US 7,693,315 B2
• Time in Patent Office: Days
• US Class Current: 600/407
• CPC Class Codes: A61B 6/507 for determination of haemodynamic parameters, e.g. perfusion CT
• CPC Class Codes: G06T 2207/30048 Heart; Cardiac
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

Systems and methods are provided for automated assessment of regional myocardial function using wall motion analysis methods that analyze various features/parameters of patient information (image data and non-image data) obtained from medical records of a patient. For example, a method for providing automatic diagnostic support for cardiac imaging generally comprises obtaining image data of a heart of a patient, obtaining features from the image data of the heart, which are related to motion of the myocardium of the heart, and automatically assessing regional myocardial function of one or more regions of a myocardial wall using the obtained features.

Claims

1. A method for providing automatic diagnostic support for cardiac imaging, comprising:

obtaining information from image data of a heart of a patient;

obtaining information from clinical data records of the patient; and

automatically assessing regional myocardial function of the heart through a wall motion analysis using the obtained information.

2. The method of claim 1, wherein obtaining information from image data comprises automatically extracting myocardial wall motion data and myocardial wall thickening data from the image data.

3. The method of claim 2, wherein the image data comprises cardiac ultrasound image data.

4. The method of claim 1, wherein obtaining information from clinical data records comprises automatically extracting clinical data from structured and/or unstructured data sources comprising the clinical data records of the patient.

5. The method of claim 1, wherein automatically assessing regional myocardial function comprises automatically assessing a condition of myocardial tissue of one or more regions of a myocardial wall using the obtained information.

6. The method of claim 1, wherein automatically assessing regional myocardial function comprises generating a wall motion score for one or more regions of a myocardial wall, wherein the wall motion score is based on a standardized scoring scheme.

7. The method of claim 6, wherein the standard scoring scheme is specified by the ASE (American Society of Echocardiography).

8. The method of claim 1, further comprising automatically determining a measure of confidence for an assessment of regional myocardial function.

9. The method of claim 1, wherein automatically assessing regional myocardial function of the heart comprises implementing a method for classifying regional myocardial function that is trained to analyze wall motion using the obtained information.

10. The method of claim 9, wherein the method for classifying regional myocardial function comprises a machine learning method, a model-based method, or any combination of machine learning and model-based methods.

11. The method of claim 9, further comprising retraining the method of classifying regional myocardial function on a continual or periodic basis using expert data and/or data learned from a plurality of case studies.

12. A method for providing automatic diagnostic support for cardiac imaging, comprising:

obtaining image data of a heart;

obtaining features from the image data of the heart, wherein the features comprise features that are related to motion of the myocardium of the heart; and

automatically assessing a condition of myocardial tissue of one or more regions of a myocardial wall using the obtained features.

13. The method of claim 12, wherein obtaining features from the image data of the heart which are related to motion of the myocardium of the heart comprises:

obtaining myocardial wall motion data from image data of a heart; and

obtaining myocardial wall thickening data from the image data of the heart.

14. The method of claim 12, wherein the image data comprises MR (magnetic resonance) image data.

15. The method of claim 12, wherein the image data comprises CT (computed tomography) image data.

16. The method of claim 12, wherein the image data comprises ultrasound image data.

17. The method of claim 16, wherein the ultrasound image data comprises image data acquired in three dimensions.

18. The method of claim 12, wherein obtaining features from the image data of the heart comprises obtaining features related to myocardial perfusion and/or obtaining features from image data of a coronary artery tree

19. The method of claim 12, wherein the myocardial wall comprises an endocardial wall of a left ventricle of the heart.

20. The method of claim 12, wherein automatically assessing comprises automatically classifying the condition of myocardial tissue of one or more regions of the myocardial wall using a method that is trained to assess myocardial function based on the obtained features.

21. The method of claim 20, wherein automatically classifying is performed using a machine learning method, a model-based method, or any combination of machine learning and model-based methods.

22. The method of claim 20, wherein obtaining features comprises obtaining features for each of a plurality of segments of a myocardial wall and wherein automatically classifying the condition of myocardial tissue of one or more regions of the myocardial wall comprises automatically classifying a condition of myocardial tissue for each segment of said plurality of segments of the myocardial wall.

23. The method of claim 22, wherein automatically classifying a condition of myocardial tissue for each segment of said plurality of segments of the myocardial wall comprises generating an indicator that indicates whether the myocardial tissue for each segment is normal or abnormal.

24. The method of claim 22, wherein automatically classifying a condition of myocardial tissue for each segment of said plurality of segments of the myocardial wall comprises generating a wall motion score for each segment based on a standard specified by the ASE (American Society of Echocardiography).

25. The method of claim 22, further comprising automatically determining a measure of confidence for each classified condition of myocardial tissue for each segment of said plurality of segments of the myocardial wall.

26. The method of claim 12, further comprising obtaining a global parameter from the image data of the heart, which provides a global indicator of heart function, and wherein the step of automatically assessing is performed using the using the obtained features and the obtained global parameter.

27. The method of claim 19, wherein the global parameter comprises left ventricular volume, left ventricular ejection fraction, left ventricular wall thickness, left ventricular wall mass, or diastolic function indicators such as the E/A ratio.

28. The method of claim 12, further comprising obtaining one or more regional parameters from the image data of the heart, including tissue velocity data, strain data, strain rate data, perfusion data, or timing data, and wherein the step of automatically assessing is performed using the using the obtained features and the obtained one or more regional parameters.

29. The method of claim 12, further comprising obtaining clinical data from clinical data records of the patient, and wherein the step of automatically assessing is performed using the using the obtained features and the obtained clinical data.

30. The method of claim 12, further comprising automatically diagnosing a medical condition using results obtained from the step of automatically assessing.

31. The method of claim 23, wherein automatically diagnosing a medical condition is performed using a classification method.

32. The method of claim 12, wherein automatically assessing a condition of myocardial tissue comprises automatically determining a probability of diagnosis of a heart disease or condition or automatically determining a probability of developing a heart disease or condition in the future.

33. The method of claim 32, wherein automatically determining a probability of diagnosis of a heart disease or condition further comprises automatically determining one or more additional features that would increase a confidence of said probability of diagnosis.

34. The method of claim 33, wherein automatically determining one or more additional features further comprises determining for each of said one or more additional features, a measure of usefulness in increasing said confidence of diagnosis.

35. The method of claim 32, wherein automatically determining a probability of diagnosis of a heart disease or condition further comprises automatically determining one or more additional cardiac imaging tests that would increase a confidence of said probability of diagnosis.

36. The method of claim 35, wherein automatically determining one or more additional cardiac imaging tests further comprises determining for each of said one or more additional tests, a measure of usefulness in increasing said confidence of diagnosis.

37. The method of claim 12, wherein automatically assessing a condition of myocardial tissue comprises automatically identifying one or more previously diagnosed cases that are similar to the current case.

38. The method of claim 37, comprising displaying the one or more identified similar cases.