| CPC G06T 5/70 (2024.01) [G06T 2207/10116 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] | 18 Claims |
|
1. A method for noise reduction in a low-dose X-ray image, acquired using a lower dose than a reference dose, using a noise reduction algorithm that includes a preprocessing step for determining input data, at least one trained function for determining noise-reduced output data from the input data, and a postprocessing step for determining a result image from the output data, the method comprising:
receiving or determining at least one result parameter specifying at least one desired result attribute of the result image that is to be achieved by the noise reduction; and
obtaining the at least one desired result attribute, obtaining the at least one desired results attribute comprising:
modifying the preprocessing, such that a noise value of at least one first noise parameter is set, the at least one first noise parameter specifying a first noise attribute, the noise value being determined from one or more of the at least one result parameters, wherein the noise value is selectable to differ from a reference value of the at least one first noise parameter, the reference value specifying a first noise attribute defined for the training data for training the trained function;
setting, according to one or more of the at least one result parameters, the at least one trained function to one of a plurality of predefined noise values of at least one second noise parameter that specifies a second noise attribute, that were set for training data used to train the at least one trained function, wherein the one noise value of the at least one second noise parameter for setting the at least one trained function is selectable to differ from a noise value of the at least one second noise parameter that prevails in the low-dose X-ray image.
|