Image processing device and image processing method

An image processing apparatus for creating a tomographic image after noise removal processing based on list data collected by a radiation tomography apparatus, the apparatus comprising:

• a first image creation unit configured to divide the list data into a plurality of frames in a collection order, and perform reconstruction processing for each of the plurality of frames using a data group in the list data included in the frame to create a first tomographic image of the frame;
• a second image creation unit configured to perform reconstruction processing using a data group in the list data having a data amount larger than that of the data group used in creating the first tomographic image of each frame in the first image creation unit to create a second tomographic image;
• and
• a CNN processing unit configured to generate a plurality of output tomographic images by, for each of the plurality of frames, repeatedly inputting the second tomographic image to a convolutional neural network, outputting an output tomographic image from the convolutional neural network, and training the convolutional neural network based on a comparison between the output tomographic image and the first tomographic image, wherein
• the second image creation unit is configured to use, in creating the second tomographic image corresponding to the first tomographic image of each frame, a data group before and after the data group used in creating the first tomographic image of the frame in the first image creation unit.

An image processing method for creating a tomographic image after noise removal processing based on list data collected by a radiation tomography apparatus, the method comprising:

• performing a first image creation of dividing the list data into a plurality of frames in a collection order, and performing reconstruction processing for each of the plurality of frames using a data group in the list data included in the frame to create a first tomographic image of the frame;
• performing a second image creation of performing reconstruction processing using a data group in the list data having a data amount larger than that of the data group used in creating the first tomographic image of each frame in the first image creation to create a second tomographic image;
• and
• performing CNN processing of generating a plurality of output tomographic images by, for each of the plurality of frames, repeatedly inputting the second tomographic image to a convolutional neural network, outputting an output tomographic image from the convolutional neural network, and training the convolutional neural network based on a comparison between the output tomographic image and the first tomographic image, wherein
• in the second image creation, in creating the second tomographic image corresponding to the first tomographic image of each frame, a data group including the data group used in creating the first tomographic image of the frame in the first image creation is used.

An image processing method for creating a tomographic image after noise removal processing based on list data collected by a radiation tomography apparatus, the method comprising:

• performing a first image creation of dividing the list data into a plurality of frames in a collection order, and performing reconstruction processing for each of the plurality of frames using a data group in the list data included in the frame to create a first tomographic image of the frame;
• performing a second image creation of performing reconstruction processing using a data group in the list data having a data amount larger than that of the data group used in creating the first tomographic image of each frame in the first image creation to create a second tomographic image;
• and
• performing CNN processing of generating a plurality of output tomographic images by, for each of the plurality of frames, repeatedly inputting the second tomographic image to a convolutional neural network, outputting an output tomographic image from the convolutional neural network, and training the convolutional neural network based on a comparison between the output tomographic image and the first tomographic image, wherein
• in the second image creation, in creating the second tomographic image corresponding to the first tomographic image of each frame, a data group before and after the data group used in creating the first tomographic image of the frame in the first image creation is used.