LEARN: Learned Experts' Assessment-based Reconstruction Network for Sparse-data CT
This work addresses a major open problem in medical imaging for CT reconstruction from sparse data, offering an incremental improvement by applying deep learning to an existing iterative method.
The paper tackles the problem of adaptive parameter selection for regularization in sparse-data CT reconstruction by proposing a Learned Experts' Assessment-based Reconstruction Network (LEARN), which achieves competitive performance on the Mayo Clinic Low-Dose Challenge Dataset in terms of artifact reduction, feature preservation, and computational speed, with only 12 layers reducing complexity by orders of magnitude.
Compressive sensing (CS) has proved effective for tomographic reconstruction from sparsely collected data or under-sampled measurements, which are practically important for few-view CT, tomosynthesis, interior tomography, and so on. To perform sparse-data CT, the iterative reconstruction commonly use regularizers in the CS framework. Currently, how to choose the parameters adaptively for regularization is a major open problem. In this paper, inspired by the idea of machine learning especially deep learning, we unfold a state-of-the-art "fields of experts" based iterative reconstruction scheme up to a number of iterations for data-driven training, construct a Learned Experts' Assessment-based Reconstruction Network ("LEARN") for sparse-data CT, and demonstrate the feasibility and merits of our LEARN network. The experimental results with our proposed LEARN network produces a competitive performance with the well-known Mayo Clinic Low-Dose Challenge Dataset relative to several state-of-the-art methods, in terms of artifact reduction, feature preservation, and computational speed. This is consistent to our insight that because all the regularization terms and parameters used in the iterative reconstruction are now learned from the training data, our LEARN network utilizes application-oriented knowledge more effectively and recovers underlying images more favorably than competing algorithms. Also, the number of layers in the LEARN network is only 12, reducing the computational complexity of typical iterative algorithms by orders of magnitude.