Spatially-Adaptive Reconstruction in Computed Tomography using Neural\n Networks

2013/11/28 by Joseph Shtok, Shtok, Joseph, Michael Zibulevsky +3
Medicine · Engineering · #Medical Imaging Techniques and Applications #Photoacoustic and Ultrasonic Imaging #Advanced X-ray and CT Imaging

paper · pdf · doi:10.48550/arxiv.1311.7251

Abstract

We propose a supervised machine learning approach for boosting existing\nsignal and image recovery methods and demonstrate its efficacy on example of\nimage reconstruction in computed tomography. Our technique is based on a local\nnonlinear fusion of several image estimates, all obtained by applying a chosen\nreconstruction algorithm with different values of its control parameters.\nUsually such output images have different bias/variance trade-off. The fusion\nof the images is performed by feed-forward neural network trained on a set of\nknown examples. Numerical experiments show an improvement in reconstruction\nquality relatively to existing direct and iterative reconstruction methods.\n

Related

• Spatially-Adaptive Reconstruction in Computed Tomography Based on Statistical Learning
• RISING a new framework for few-view tomographic image reconstruction with deep learning
• A hierarchical approach to deep learning and its application to tomographic reconstruction
• Improving the Generalisation of Learned Reconstruction Frameworks
• Accelerated Optimization of Implicit Neural Representations for CT Reconstruction