Robust multimaterial decomposition of spectral CT using convolutional neural networks

Zhengyang Chen,Liang Li

doi:10.1117/1.oe.58.1.013104

Zhengyang Chen, Liang Li

Open Access

PDF Available

https://doi.org/10.1117/1.oe.58.1.013104

Copy DOI

Export

Save

Cite

Journal: Optical Engineering	Publication Date: Jan 12, 2019
Citations: 28	License type: cc-by

Affiliation: Tsinghua University

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

Spectral computed tomography (CT) can reconstruct scanned objects at different energy-bins and thus solve the multimaterial decomposition (MMD) problem. Because the linear attenuation coefficients of different basis materials may be extremely close, the decomposition problem is often ill-conditioned. Meanwhile, traditional material decompositions with image-domain algorithms are usually voxelwise based. Therefore, these algorithms rely heavily on image quality. Ring artifacts often exist in the reconstructed images of spectral CT due to the inconsistency feature of energy-resolved detectors and beam-hardening effect. Considering the enlargement of the receptive field and taking advantage of the modeling ability of convolutional neural networks in deep learning, we proposed a convolutional material decomposition algorithm to solve the MMD problem through a basis of patches instead of pixels of the spectral CT images. Simulations and physical experiments were performed to validate the proposed algorithm, and its quality was compared with a traditional MMD algorithm in the image domain. Results show that the proposed method achieves good accuracy, reduces mean squared errors by one to two orders, and exhibits robustness in the MMD of spectral CT images even in the case that obvious ring artifacts is presented.

Highlights

The ability of spectral computed tomography (CT) to distinguish photons in terms of their different energies allows this technique to decompose scanned objects into their basis materials.[1,2,3,4,5] Compared with linear attenuation coefficients from traditional single energy CT, the distribution of basis materials can effectively reflect inner information
Image-domain methods assume that the effective attenuation coefficients of each pixel of the spectral CT images can be decomposed into the linear combination of several basis functions.[10,11,12,13,14,15,16]
The performance of image-domain method is usually affected by the ring artifacts and beam-hardening artifacts within the CT images especially for the spectral CT using the photon-counting detectors

Summary

Introduction

The ability of spectral computed tomography (CT) to distinguish photons in terms of their different energies allows this technique to decompose scanned objects into their basis materials.[1,2,3,4,5] Compared with linear attenuation coefficients from traditional single energy CT, the distribution of basis materials can effectively reflect inner information. Ref. 22 uses the linear attenuation curves of preselected materials as the basis functions, leading an analytical solution of MMD problem. Regardless of the choice of basis functions, these traditional material decomposition algorithms are voxelwise based, using no more than dozens of pixels around the target pixel Their performances usually depend on the reconstructed image quality of spectral CT. An idea to avoid these problems is using the data-driven methods, for example, machine learning, instead of the analytical solution such as Ref. 22 In this way, we can expand the receptive field of data-driven algorithms to provide more information for their learning

Methods

Results

Conclusion