System and method for determining optimal operating parameters for medical imaging

• US 10,074,037 B2
• Filed: 06/03/2016
• Issued: 09/11/2018
• Est. Priority Date: 06/03/2016
• Status: Active Grant

• Current Assignee: Siemens Healthineers AG (Siemens AG)
• Original Assignee: Siemens Healthcare GMBH (Siemens AG)
• Inventors: Lu, Xiaoguang, Deshpande, Vibhas, Kollasch, Peter, Wang, Dingxin, Sharma, Puneet
• Primary Examiner(s): Bitar, Nancy
• Application Number:
  US15/172,949
• Publication Number:
  US 20170351937A1
• Time in Patent Office: 830 Days
• Field of Search: None
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: A61B 5/4585 Evaluating the knee
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/037 Emission tomography
• CPC Class Codes: A61B 8/58 Testing, adjusting or calibrating the diagnostic device
• CPC Class Codes: G01R 33/543 Control of the operation of the MR system, e.g. setting of acquisition parameters prior to or during MR data acquisition, dynamic shimming, use of one or more scout images for scan plane prescription G01R33/546 takes precedence
• CPC Class Codes: G06F 18/2193 based on specific statistical tests
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06T 2207/30048 Heart; Cardiac
• CPC Class Codes: G06T 2207/30101 Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/0014 using an image reference approach

Abstract

Systems and methods for determining optimized imaging parameters for imaging a patient include learning a model of a relationship between known imaging parameters and a quality measure, the known imaging parameters and the quality measure being determined from training data. Optimized imaging parameters are determined by optimizing the quality measure using the learned model. Images of the patient are acquired using the optimized imaging parameters.

Claims

1. A method for determining optimized imaging parameters for imaging a patient, comprising:

learning a model of a relationship between known imaging parameters and a quality measure, the known imaging parameters and the quality measure being determined from training data, wherein the quality measure is based on a signal-to-noise ratio (SNR) determined as a ratio between an average intensity value of pixels in a first region of interest in a target structure in an image and an average intensity value of pixels in a second region of interest outside of the target structure in the image;

determining optimized imaging parameters by optimizing the quality measure using the learned model, the optimized imaging parameters comprising parameters of an image acquisition device and patient parameters for imaging the patient, the optimized imaging parameters being constrained to be within a range of the known imaging parameters; and

acquiring images of the patient using the image acquisition device based on the optimized imaging parameters.

2. The method as recited in claim 1, wherein optimizing the quality measure using the learned model comprises maximizing the quality measure.

3. The method as recited in claim 1, wherein the quality measure is further based on an additional metric, the additional metric comprising at least one of a contrast-to-noise ratio, a resolution, a measure of motion artifacts, and a measure of ghosting.

4. The method as recited in claim 3, wherein the additional metric is selected based on a procedure being performed.

5. The method as recited in claim 1, wherein the known imaging parameters comprise parameters for operating the image acquisition device.

6. The method as recited in claim 1, wherein optimizing the quality measure comprises optimizing the quality measure using an interior point algorithm.

7. The method as recited in claim 1, wherein learning the model comprises learning the model using a regression analysis.

8. An apparatus for determining optimized imaging parameters for imaging a patient, comprising:

means for learning a model of a relationship between known imaging parameters and a quality measure, the known imaging parameters and the quality measure being determined from training data, wherein the quality measure is based on a signal-to-noise ratio (SNR) determined as a ratio between an average intensity value of pixels in a first region of interest in a target structure in an image and an average intensity value of pixels in a second region of interest outside of the target structure in the image;

means for determining optimized imaging parameters by optimizing the quality measure using the learned model, the optimized imaging parameters comprising parameters of an image acquisition device and patient parameters for imaging the patient, the optimized imaging parameters being constrained to be within a range of the known imaging parameters; and

means for acquiring images of the patient using the optimized imaging parameters.

9. The apparatus as recited in claim 8, wherein the means for determining optimized imaging parameters by optimizing the quality measure using the learned model comprises means for maximizing the quality measure.

10. The apparatus as recited in claim 8, wherein the quality measure is further based on an additional metric, the additional metric comprising at least one of a contrast-to-noise ratio, a resolution, a measure of motion artifacts, and a measure of ghosting.

11. The apparatus as recited in claim 10, wherein the additional metric is selected based on a procedure being performed.

12. The apparatus as recited in claim 8, wherein the known imaging parameters comprise parameters for operating the image acquisition device.

13. The apparatus as recited in claim 8, wherein the means for optimizing the quality measure comprises means for optimizing the quality measure using an interior point algorithm.

14. The apparatus as recited in claim 8, wherein the means for learning the model comprises means for learning the model using a regression analysis.

15. A non-transitory computer readable medium storing computer program instructions for determining optimized imaging parameters for imaging a patient, the computer program instructions when executed by a processor cause the processor to perform operations comprising:

learning a model of a relationship between known imaging parameters and a quality measure, the known imaging parameters and the quality measure being determined from training data, wherein the quality measure is based on a signal-to-noise ratio (SNR) determined as a ratio between an average intensity value of pixels in a first region of interest in a target structure in an image and an average intensity value of pixels in a second region of interest outside of the target structure in the image;

determining optimized imaging parameters by optimizing the quality measure using the learned model, the optimized imaging parameters comprising parameters of an image acquisition device and patient parameters for imaging the patient, the optimized imaging parameters being constrained to be within a range of the known imaging parameters; and

acquiring images of the patient using the optimized imaging parameters.

16. The non-transitory computer readable medium as recited in claim 15, wherein optimizing the quality measure using the learned model comprises maximizing the quality measure.

17. The non-transitory computer readable medium as recited in claim 15, wherein the quality measure is further based on an additional metric, the additional metric comprising at least one of a contrast-to-noise ratio, a resolution, a measure of motion artifacts, and a measure of ghosting.

18. The non-transitory computer readable medium as recited in claim 17, wherein the additional metric is selected based on a procedure being performed.

19. The non-transitory computer readable medium as recited in claim 15, wherein the known imaging parameters comprise parameters for operating the image acquisition device.

20. The non-transitory computer readable medium as recited in claim 15, wherein optimizing the quality measure comprises optimizing the quality measure using an interior point algorithm.

21. The non-transitory computer readable medium as recited in claim 15, wherein learning the model comprises learning the model using a regression analysis.