Deep Learning-Guided Image Reconstruction from Incomplete Data

Abstract

An approach to incorporate deep learning within an iterative image reconstruction framework to reconstruct images from severely incomplete measurement data is presented. Specifically, we utilize a convolutional neural network (CNN) as a quasi-projection operator within a least squares minimization procedure. The CNN is trained to encode high level information about the class of images being imaged; this information is utilized to mitigate artifacts in intermediate images produced by use of an iterative method. The structure of the method was inspired by the proximal gradient descent method, where the proximal operator is replaced by a deep CNN and the gradient descent step is generalized by use of a linear reconstruction operator. It is demonstrated that this approach improves image quality for several cases of limited-view image reconstruction and that using a CNN in an iterative method increases performance compared to conventional image reconstruction approaches. We test our method on several limited-view image reconstruction problems. Qualitative and quantitative results demonstrate state-of-the-art performance.

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections
• Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures
• ReconNet: Non-Iterative Reconstruction of Images from Compressively Sensed Random Measurements

Cited by in corpus (8)

• Model based learning for accelerated, limited-view 3D photoacoustic tomography
• Real-time photoacoustic projection imaging using deep learning
• Deep learning versus ${\ell }^1$-minimization for compressed sensing photoacoustic tomography
• Solving Linear Inverse Problems Using GAN Priors: An Algorithm with Provable Guarantees
• NETT Regularization for Compressed Sensing Photoacoustic Tomography
• Random mesh projectors for inverse problems
• Survey of Deep Learning Methods for Inverse Problems
• Learned backprojection for sparse and limited view photoacoustic tomography