Deep Kernel Representation for Image Reconstruction in PET
This work addresses image reconstruction in PET, a domain-specific problem for medical imaging, with incremental improvements over prior kernel methods.
The authors tackled the challenge of PET image reconstruction by proposing a deep kernel method that uses a neural network to learn improved kernels automatically, outperforming existing kernel and neural network methods in simulations and real patient data.
Image reconstruction for positron emission tomography (PET) is challenging because of the ill-conditioned tomographic problem and low counting statistics. Kernel methods address this challenge by using kernel representation to incorporate image prior information in the forward model of iterative PET image reconstruction. Existing kernel methods construct the kernels commonly using an empirical process, which may lead to unsatisfactory performance. In this paper, we describe the equivalence between the kernel representation and a trainable neural network model. A deep kernel method is then proposed by exploiting a deep neural network to enable automated learning of an improved kernel model and is directly applicable to single subjects in dynamic PET. The training process utilizes available image prior data to form a set of robust kernels in an optimized way rather than empirically. The results from computer simulations and a real patient dataset demonstrate that the proposed deep kernel method can outperform the existing kernel method and neural network method for dynamic PET image reconstruction.