Systems and methods providing automated decision support and medical imaging

• US 7,672,491 B2
• Filed: 03/17/2005
• Issued: 03/02/2010
• Est. Priority Date: 03/23/2004
• Status: Active Grant

• Current Assignee: Siemens Healthineers AG (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Cannon, Michael G., Dutton, Allen, Krishnan, Sriram, Zhou, Xiang Sean, Comaniciu, Dorin
• Primary Examiner(s): AZARIAN, SEYED H
• Application Number:
  US11/083,161
• Publication Number:
  US 20050251013A1
• Time in Patent Office: 1,811 Days
• Field of Search: 382/100, 382/128, 382/129, 382/130, 382/131, 382/132, 382/155, 382/157, 382/168, 382/170, 382/181, 382/203, 382/224, 382/232, 382/260, 382/274, 382/276, 382/287, 382/133, 382/305, 382/133.155, 600/300, 600/425, 600/518, 378/21
• US Class Current: 382/128
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

Systems and methods are provided for processing a medical image to automatically identify the anatomy and view (or pose) from the medical image and automatically assess the diagnostic quality of the medical image. In one aspect a method for automated decision support for medical imaging includes obtaining image data, extracting feature data from the image data, and automatically performing anatomy identification, view identification and/or determining a diagnostic quality of the image data, using the extracted feature data.

Claims

1. A computer readable medium embodying instructions executed by a processor to perform a method for providing automated decision support for medical imaging, the method comprising:

obtaining an image comprising image data;

extracting feature data from the image data; and

automatically assessing an error in the image using the extracted feature data, wherein the error corresponds to a diagnostic quality on the image;

wherein automatically assessing an error is performed by using the extracted feature data to query a database of templates derived from information of known cases to identify similar templates and use information associated with the identified similar templates to determine the diagnostic quality.

2. The method of claim 1, wherein the image data comprises cardiac ultrasound image data.

3. The method of claim 1, further comprising automatically determining a metric that indicates a level of assessed diagnostic quality, wherein the error is a deviation in the image from an expected image.

4. The method of claim 1, wherein automatically assessing image quality is performed by processing the extracted feature data using a classifier to determine the diagnostic quality.

5. The method of claim 4, wherein the classifier is implemented using a machine learning method, a model-based method, or any combination of machine learning and model-based methods.

6. The method of claim 1, wherein automatically assessing diagnostic quality is performed by using the extracted feature data to query a database of known cases to identify similar cases and use information associated with the identified similar cases to determine the diagnostic quantity.

7. The method of claim 1, further comprising automatically identifying and labeling an anatomical object of the image data using the extracted feature data.

8. The method of claim 1, wherein automatically assessing diagnostic quality of the image further comprises using results of automatic anatomy identification.

9. The method of claim 1, further comprising automatically identifying and labeling a view of the image using the extracted feature data.

10. The method of claim 9, wherein automatically assessing diagnostic quality of the image further comprises using results of automatic view identification.

11. The method of claim 9, wherein the view is a standard view for a clinical domain of interest.

12. The method of claim 1, wherein automatically assessing diagnostic quality of the image using the extracted feature data is performed in real-time during image acquisition.

13. The method of claim 12, further comprising presenting assessed diagnostic quality results to a user in real-time during image acquisition.

14. A computer readable medium embodying instructions executed by a processor to perform a method for providing automated decision support for medical imaging, the method comprising:

obtaining an image comprising image data;

extracting feature data from the image data;

performing an automatic anatomy, identification process using the extracted feature data to label an identified anatomical object in the image data;

performing an automatic view identification process using the extracted feature data and results of anatomy identification to label a view of the image data; and

performing an automated image quality assessment process using the extracted feature data and results of anatomy and view identification to assess a level of diagnostic quality of the image, wherein the diagnostic quality corresponds to error introduced in an image acquisition process;

wherein automatically assessing an error is performed by using the extracted feature data to query a database of templates derived from information of known cases to identify similar templates and use information associated with the identified similar templates to determine the diagnostic quality.

15. The method of claim 14, wherein automated anatomy identification, view identification and diagnostic quality assessment are performed in real-time during image acquisition.

16. The method of claim 15, further comprising presenting results of the automated anatomy identification, view identification and diagnostic quality assessment processes to a user in real-time during image acquisition.

17. The method of claim 14, wherein automated anatomy identification, view identification and diagnostic quality assessment are performed using associated classifiers that process the extracted feature data.

18. The method of claim 17, wherein the classifiers are implemented using a machine learning method a model-based method, or any combination of machine learning and model-based methods.

19. The method of claim 14, wherein automated anatomy identification view identification and diagnostic quality assessment are performed by using the extracted feature data to query a database of known cases to identify similar cases and use information associated with the identified similar cases to perform the respective functions.

20. The method of claim 14, wherein automated anatomy identification view identification and diagnostic quality assessment are performed by using the extracted feature data to query a database of templates derived from information of known cases to identify similar templates and use information associated with the identified similar templates to perform the respective functions.

21. A computer readable medium embodying instructions executed by a processor to perform a method for providing automated decision support for medical imaging, the method comprising:

obtaining an image comprising image data;

extracting feature data from the image data; and

performing an automatic view identification process using the extracted feature data to identify a view of the image; and

performing an automated image quality assessment process using the extracted feature data and results of view identification to assess a level of diagnostic quality of the image, wherein the diagnostic quality corresponds to error introduced in an image acquisition process;

wherein automatically assessing an error is performed by using the extracted feature data to query a database of templates derived from information of known cases to identify similar templates and use information associated with the identified similar templates to determine the diagnostic quality.

22. The method of claim 21, wherein the image data comprises image data of a heart acquired using ultrasound imaging.

23. The method of claim 22, wherein a view is identified as an apical two-chamber view (A2C), an apical four-chamber view (A4C) an apical long axis view (ALAX), the parasternal long axis view (PLAX), or a parasternal short axis view (PSAX).

24. The method of claim 21, wherein the automated diagnostic quality assessment process includes evaluating the image to identify a foreshortened myocardium apex, wherein the image is a cardiac image.

25. The method of claim 24, wherein automatically determining diagnostic quality of the image using the extracted feature data is performed in real-time during image acquisition.

26. The method of claim 25, further comprising presenting diagnostic quality determination results to a user in real-time during image acquisition, wherein the error is a deviation in the image from an expected image.