Method and system for generating multi-task learning-type generative adversarial network for low-dose pet reconstruction

A system for generating a multi-task learning-type generative adversarial network for low-dose positron emission tomography (PET) reconstruction, comprising:

• a picture generator acquiring module, configured to connect layers of an encoder and layers of a decoder by skip connection to provide a U-Net type picture generator;
• an generative adversarial network generation module, configured to match a plurality of picture generators with a plurality of discriminators by one-to-one manner to provide a group of generative adversarial networks, wherein the plurality of picture generators use an input modality as a conditional input and generating desired PET images as a learning objective, the plurality of discriminators use an input modality of a corresponding picture generator, a tag image corresponding to the input modality, and an output result as an input, and, in each group of the generative adversarial network, the input modality comprises at least a low-dose PET image and an magnetic resonance (MR) image of a same picture object;
• a first learning-type generative adversarial network acquiring module, configured to leave the plurality of generative adversarial networks in the group of the generative adversarial networks to learn in parallel and leave the picture generators of all the generative adversarial networks to share shallow information to provide a first multi-task learning-type generative adversarial network;
• a joint loss function acquiring module, configured to evaluate parameters of the first multi-task learning-type generative adversarial network by using a standard-dose PET picture as a tag picture corresponding to the input modality and using L1 type loss function and cross entropy loss function, and design a joint loss function 1 for improving image quality according to output results of the picture generators, the tag picture and the output results of the discriminators;
• and
• a second multi-task leaning-type generative adversarial network acquiring module, configured to train the first multi-task learning-type generative adversarial network according to the joint loss function 1 in combination with an optimizer to provide a second multi-task learning-type generative adversarial network.