Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep Learning Computed Tomography based on the Defrise and Clack Algorithm

Chengze Ye, Linda-Sophie Schneider, Yipeng Sun, Andreas Maier

TL;DR

This study presents a novel approach for reconstructing cone beam computed tomography (CBCT) for specific orbits using known operator learning using a filtered backprojection type (FBP-type) algorithm, which integrates a unique, adaptive filtering process.

Abstract

This study presents a novel approach for reconstructing cone beam computed tomography (CBCT) for specific orbits using known operator learning. Unlike traditional methods, this technique employs a filtered backprojection type (FBP-type) algorithm, which integrates a unique, adaptive filtering process. This process involves a series of operations, including weightings, differentiations, the 2D Radon transform, and backprojection. The filter is designed for a specific orbit geometry and is obtained using a data-driven approach based on deep learning. The approach efficiently learns and optimizes the orbit-related component of the filter. The method has demonstrated its ability through experimentation by successfully learning parameters from circular orbit projection data. Subsequently, the optimized parameters are used to reconstruct images, resulting in outcomes that closely resemble the analytical solution. This demonstrates the potential of the method to learn appropriate parameters from any specific orbit projection data and achieve reconstruction. The algorithm has demonstrated improvement, particularly in enhancing reconstruction speed and reducing memory usage for handling specific orbit reconstruction.

Deep Learning Computed Tomography based on the Defrise and Clack Algorithm

TL;DR

This study presents a novel approach for reconstructing cone beam computed tomography (CBCT) for specific orbits using known operator learning using a filtered backprojection type (FBP-type) algorithm, which integrates a unique, adaptive filtering process.

Abstract

This study presents a novel approach for reconstructing cone beam computed tomography (CBCT) for specific orbits using known operator learning. Unlike traditional methods, this technique employs a filtered backprojection type (FBP-type) algorithm, which integrates a unique, adaptive filtering process. This process involves a series of operations, including weightings, differentiations, the 2D Radon transform, and backprojection. The filter is designed for a specific orbit geometry and is obtained using a data-driven approach based on deep learning. The approach efficiently learns and optimizes the orbit-related component of the filter. The method has demonstrated its ability through experimentation by successfully learning parameters from circular orbit projection data. Subsequently, the optimized parameters are used to reconstruct images, resulting in outcomes that closely resemble the analytical solution. This demonstrates the potential of the method to learn appropriate parameters from any specific orbit projection data and achieve reconstruction. The algorithm has demonstrated improvement, particularly in enhancing reconstruction speed and reducing memory usage for handling specific orbit reconstruction.
Paper Structure (11 sections, 13 equations, 4 figures, 1 table)

This paper contains 11 sections, 13 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Filtered Backprojection
  4. Grangeat's Inversion
  5. Defrise and Clack Algorithm
  6. Defrise and Clack Neural Network
  7. Grangeat's intermediate function
  8. Filter the cone-beam data
  9. Backprojection to the 3D image
  10. Experiments and Results
  11. Conclusion and Discussion

Figures (4)

  • Figure 1: Grangeat's intermediate function part of the neural network.
  • Figure 2: Filtering part of the neural network.
  • Figure 3: Grangeat's intermediate function part, filtering part, and backprojection part are combined to form the Defrise and Clack neural network architecture.
  • Figure 4: Reconstructed results of the network. (a) Reconstruction using learned redundancy weight. (b) Reconstruction using analytic redundancy weight. (c) FDK reconstruction result.