Synthesis of Positron Emission Tomography (PET) Images via Multi-channel Generative Adversarial Networks (GANs)
This work addresses the need for boosting training data in computer-aided diagnosis systems for medical imaging, specifically for PET image synthesis, but it appears incremental as it builds on existing GAN approaches with multi-channel inputs.
The paper tackled the problem of synthesizing low-resolution PET images by proposing a multi-channel GAN (M-GAN) method, which resulted in synthetic PET images that were much closer to real images compared to existing methods, as demonstrated on 50 lung cancer PET-CT studies.
Positron emission tomography (PET) image synthesis plays an important role, which can be used to boost the training data for computer aided diagnosis systems. However, existing image synthesis methods have problems in synthesizing the low resolution PET images. To address these limitations, we propose multi-channel generative adversarial networks (M-GAN) based PET image synthesis method. Different to the existing methods which rely on using low-level features, the proposed M-GAN is capable to represent the features in a high-level of semantic based on the adversarial learning concept. In addition, M-GAN enables to take the input from the annotation (label) to synthesize the high uptake regions e.g., tumors and from the computed tomography (CT) images to constrain the appearance consistency and output the synthetic PET images directly. Our results on 50 lung cancer PET-CT studies indicate that our method was much closer to the real PET images when compared with the existing methods.