Imaging method and apparatus with optimized grayscale value window determination

1. A tomographic imaging method, which comprises the following steps:

displaying a plurality of training data sets on a display apparatus;

selecting a grayscale value setting for each training data set that should be displayed on the display apparatus with the grayscale value setting;

assigning a feature set with at least one feature to each training data set;

storing the grayscale value setting selected for the training data set and the feature set assigned to the training data set, for each of the plurality of training data sets;

assigning a feature set with at least one feature to an examination data set;

automatically selecting the grayscale value setting for the examination data set depending on the feature set of the examination data set and depending on the feature set of the training data sets;

displaying the examination data set with the automatically selected grayscale value setting on the display apparatus; and

generating a transformation between a feature set of the examination data set and the grayscale value settings for the examination data set by interpolating between the grayscale value settings for a plurality of training data sets;

wherein the transformation between the feature set of the examination data set and the grayscale value settings for the examination data set is determined by at least one of the following steps:

using a regression analysis;

using a linear regression model;

using a simple linear regression;

using a logistic regression;

using a nonlinear regression;

using a piecewise regression; and

using a nearest-neighbor model.

2. The imaging method according to claim 1 , wherein the training data sets display the same tissue regions, created with the same modality, of a plurality of different patients.

3. The imaging method according to claim 1 , which comprises:

extracting at least one feature of the feature set from the training data sets; and

extracting at least one feature of the feature set from the examination data set.

4. The imaging method according to claim 1 , which further comprises carrying out one or more of the following steps:

determining a tissue density from the training data sets and the examination data set as one feature of the feature set;

determining a component of fatty tissue from the training data sets and the examination data set as one feature of the feature set;

determining a component of glandular tissue from the training data sets and the examination data set as one feature of the feature set;

determining a component of scarred tissue from the training data sets and the examination data set as one feature of the feature set;

determining a component of bright tissue from the training data sets and the examination data set as one feature of the feature set; and

determining a component of calcified tissue from the training data sets and the examination data set as one feature of the feature set.

5. The imaging method according to claim 1 , which further comprises assigning a radiation dose to each of the training data sets and the examination data set as a feature of the feature set.

6. The imaging method according to claim 1 , which further comprises the following steps:

determining the training data set whose feature set has a greatest correspondence to the examination data set; and

displaying the examination data set with a grayscale value setting that is assigned to the training data set whose feature set has the greatest correspondence to the examination data set.

7. The imaging method according to claim 1 , which further comprises the following steps:

modifying the grayscale value setting of the displayed examination data set to define a modified grayscale value setting; and

storing the modified grayscale value setting and the feature set of the examination data set as the grayscale value setting and the feature set of a training data set.

8. The imaging method according to claim 1 , wherein the grayscale value setting comprises at least one of the following:

a grayscale value range of the examination data set to be displayed on a display apparatus;

a contrast of the displayed image;

a mean brightness of the displayed image; and

parameters of a false-color representation of the displayed image.

9. A representation learning apparatus, configured to carry out the following method steps:

displaying a plurality of training data sets on a display apparatus;

selecting a grayscale value setting for each training data set that should be displayed on the display apparatus with the grayscale value setting;

assigning a feature set with at least one feature to each training data set;

storing the grayscale value setting selected for the training data set and the feature set assigned to the training data set, for each of the plurality of training data sets;

assigning a feature set with at least one feature to an examination data set;

automatically selecting the grayscale value setting for the examination data set depending on the feature set of the examination data set and depending on the feature set of the training data sets;

displaying the examination data set with the automatically selected grayscale value setting on the display apparatus; and

generating a transformation between a feature set of the examination data set and the grayscale value settings for the examination data set by interpolating between the grayscale value settings for a plurality of training data sets;

wherein the transformation between the feature set of the examination data set and the grayscale value settings for the examination data set is determined by at least one of the following steps:

using a regression analysis;

using a linear regression model;

using a simple linear regression;

using a logistic regression;

using a nonlinear regression;

using a piecewise regression; and

using a nearest-neighbor model.

10. An imaging system, comprising a representation learning apparatus for carrying out the following method steps:

displaying a plurality of training data sets on a display apparatus;

selecting a grayscale value setting for each training data set that should be displayed on the display apparatus with the grayscale value setting;

assigning a feature set with at least one feature to each training data set;

storing the grayscale value setting selected for the training data set and the feature set assigned to the training data set, for each of the plurality of training data sets;

assigning a feature set with at least one feature to an examination data set;

automatically selecting the grayscale value setting for the examination data set depending on the feature set of the examination data set and depending on the feature set of the training data sets;

displaying the examination data set with the automatically selected grayscale value setting on the display apparatus; and

generating a transformation between a feature set of the examination data set and the grayscale value settings for the examination data set by interpolating between the grayscale value settings for a plurality of training data sets;

wherein the transformation between the feature set of the examination data set and the grayscale value settings for the examination data set is determined by at least one of the following steps:

using a regression analysis;

using a linear regression model;

using a simple linear regression;

using a logistic regression;

using a nonlinear regression;

using a piecewise regression; and

using a nearest-neighbor model.

11. A non-transitory computer program product, comprising computer-executable code in non-transitory form to be loaded into, or loaded in, a memory of a computer, and configured to carry out the following executed steps on the computer:

displaying a plurality of training data sets on a display apparatus;

selecting a grayscale value setting for each training data set that should be displayed on the display apparatus with the grayscale value setting;

assigning a feature set with at least one feature to each training data set;

storing the grayscale value setting selected for the training data set and the feature set assigned to the training data set, for each of the plurality of training data sets;

assigning a feature set with at least one feature to an examination data set;

automatically selecting the grayscale value setting for the examination data set depending on the feature set of the examination data set and depending on the feature set of the training data sets;

displaying the examination data set with the automatically selected grayscale value setting on the display apparatus; and

generating a transformation between a feature set of the examination data set and the grayscale value settings for the examination data set by interpolating between the grayscale value settings for a plurality of training data sets;

wherein the transformation between the feature set of the examination data set and the grayscale value settings for the examination data set is determined by at least one of the following steps:

using a regression analysis;

using a linear regression model;

using a simple linear regression;

using a logistic regression;

using a nonlinear regression;

using a piecewise regression; and

using a nearest-neighbor model.