Tomographic image processing apparatus and method of separating material of object in spectral tomographic image, and computer program product

The tomographic image processing apparatus of claim wherein the at least one processor is further configured to:

• train a deep neural network with the augmented training data;
• generate a deep neural network model for differentiating the material separation information with respect to each pixel from the input tomographic image data of the input tomographic image;
• and
• identify the material separation information in the input tomographic image data by using the deep neural network model.

A tomographic image processing method comprising:

• displaying an input tomographic image of an object;
• determining a material combination to be separated from the input tomographic image;
• obtaining a material separation model for acquiring the material separation information based on the input tomographic image data of the input tomographic image by using training data which includes a plurality of pieces of spectral tomographic image data and is acquired when information about a proportion of each material and its energy attenuation value are known;
• generating augmented training data by fitting a concentration or density to the training data;
• and
• displaying material separation information corresponding to the determined material combination for a region of interest selected in the input tomographic image based on an external input,
• wherein the input tomographic image is a spectral tomographic image comprising a plurality of tomographic images respectively corresponding to a plurality of energy levels.

A non-transitory computer recording medium having stored therein program commands, when executed by a processor, cause the processor to perform a tomographic image processing method comprising:

• displaying an input tomographic image of an object;
• determining a material combination to be separated from the input tomographic image;
• obtaining a material separation model for acquiring the material separation information based on the input tomographic image data of the input tomographic image by using training data which includes a plurality of pieces of spectral tomographic image data and is acquired when information about a proportion of each material and its energy attenuation value are known;
• generating augmented training data by fitting a concentration or density to the training data;
• and
• displaying material separation information corresponding to the determined material combination for a region of interest selected in the input tomographic image based on an external input,
• wherein the input tomographic image is a spectral tomographic image comprising a plurality of tomographic images respectively corresponding to a plurality of energy levels.