Method for determining organ-dependent parameters for image post-processing and image processing device

• US 20060133658A1
• Filed: 12/15/2005
• Published: 06/22/2006
• Est. Priority Date: 12/17/2004
• Status: Active Grant

• Current Assignee: Siemens Healthcare GMBH (Siemens AG)
• Original Assignee: Siemens AG
• Inventors: Spahn, Martin
• Granted Patent: US 7,580,554 B2
• Time in Patent Office: Days
• US Class Current: 382/128
• CPC Class Codes: A61B 6/03 Computed tomography [CT] echo-tomography A61B8/14
• CPC Class Codes: A61B 6/5294 involving using additional data, e.g. patient information, image labeling, acquisition parameters
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 5/00 Image enhancement or restoration
• CPC Class Codes: G06T 5/60 using machine learning, e.g. neural networks

Abstract

Method for determining organ-dependent parameters for image post-processing and image processing device The invention relates to a method for the automatic determination of one or more organ-dependent parameters for image post-processing, having the following steps: recording of a raw image and the execution of image preprocessing on the raw image data in order to generate a preprocessed image, determination of an organ-specific comparison image, stored in a database, which has been optimized by postprocessing using organ-specific postprocessing parameters, and acquisition of the postprocessing parameters for the comparison image, and postprocessing of the preprocessed raw image, using the postprocessing parameters of the comparison image.

Claims

1-10. -10. (canceled)

11. A method for determining at least one organ-dependent parameter for medical image post-processing, comprising:

recording a raw image, the raw image including an organ;

applying image preprocessing to the raw image for generating a preprocessed image;

identifying an organ-specific comparison image stored in a comparison image database having a plurality of comparison images, the comparison images optimized by image postprocessing using organ-specific postprocessing parameters;

acquiring the postprocessing parameters of the organ-specific comparison image; and

postprocessing the preprocessed image using the postprocessing parameters.

12. The method according to claim 11, wherein the organ-specific comparison image is identified by comparing at least part of the preprocessed image to the comparison images.

13. The method according to claim 11, wherein the organ-specific comparison image is identified by analyzing the preprocessed image using a neural network trained by the comparison images, the analyzing generating information on the organ included in the preprocessed image.

14. The method in accordance with claim 12, wherein comparing the preprocessed image to the comparison images includes mathematically subtracting the pre-processed image from the comparison image or mathematically subtracting the comparison images from the preprocessed image.

15. The method in accordance with claim 12, wherein comparing the preprocessed image to the comparison images includes analyzing the preprocessed image for generating information on the organ included in the postprocessing image, the generated information on the organ used for identifying the organ-specific comparison image.

16. The method in accordance with claim 15, wherein

the information on the organ includes image structures present in the preprocessed image, and

the organ-specific comparison image is identified by comparing the image structures to related image structures present in the comparison images.

17. The method in accordance with claim 16, wherein the image structures are determined using a segmentation algorithm.

18. An image processing device, comprising a computing device for processing image data, the computing device configured to:

apply an image preprocessing algorithm to a raw image, the raw image including an organ, for generating a preprocessed image,

identify an organ-specific comparison image stored in a comparison image database having a plurality of comparison images, the comparison images optimized by image postprocessing using organ-specific postprocessing parameters,

acquire the postprocessing parameters of the organ-specific comparison image, and

postprocess the preprocessed image using the postprocessing parameters.

19. The image processing device according to claim 18, wherein the organ-specific comparison image is identified by comparing at least part of the preprocessed image to the comparison images.

20. The image processing device according to claim 18, further comprising a neural network trained by the comparison images, wherein the organ-specific comparison image is identified by analyzing the preprocessed image using the neural network, the analyzing generating information on the organ included in the preprocessed image.

21. The image processing device in accordance with claim 18, wherein the comparison images are stored in the comparison image database together with their related postprocessing parameters, the comparison image database configured to selectively read out the comparison images or the postprocessing parameters.

22. The image processing device in accordance with claim 19, further comprising an analysis unit configured to:

compare the preprocessed image to the comparison images,

identify the organ-specific comparison image, and

acquire the postprocessing parameters based on the comparison.

23. The image processing device in accordance with claim 22, wherein, for comparing the preprocessed image to the comparison images, the analysis unit is configured to mathematically subtract the pre-processed image from the comparison image or to mathematically subtract the comparison images from the preprocessed image.

24. The image processing device in accordance with claim 22, wherein the analysis unit is configured to:

determine image structures present in the preprocessed image, and

determine the organ-specific comparison image by comparing the image structures to related image structures present in the comparison images.