Modified Kernel MLAA Using Autoencoder for PET-enabled Dual-Energy CT

TL;DR

This paper introduces a modified kernel MLAA algorithm that employs an autoencoder CNN to extract intrinsic features from x-ray CT images, significantly enhancing GCT image quality and material decomposition in PET/CT imaging.

Contribution

The novel contribution is integrating an autoencoder CNN for feature extraction into kernel MLAA, improving noise suppression and image quality over traditional intensity-based methods.

Findings

Autoencoder kernel MLAA outperforms existing methods in image quality.

Significant noise reduction and improved material decomposition.

Potential over-smoothness in bone regions noted.

Abstract

Combined use of PET and dual-energy CT provides complementary information for multi-parametric imaging. PETenabled dual-energy CT combines a low-energy x-ray CT image with a high-energy &\gamma&-ray CT (GCT) image reconstructed from time-of-flight PET emission data to enable dual-energy CT material decomposition on a PET/CT scanner. The maximumlikelihood attenuation and activity (MLAA) algorithm has been used for GCT reconstruction but suffers from noise. Kernel MLAA exploits an x-ray CT image prior through the kernel framework to guide GCT reconstruction and has demonstrated substantial improvements in noise suppression. However, similar to other kernel methods for image reconstruction, the existing kernel MLAA uses image intensity-based features to construct the kernel representation, which is not always robust and may lead to suboptimal reconstruction with artifacts. In this paper, we propose a modified kernel method by using an autoencoder convolutional neural network (CNN) to extract an intrinsic feature set from the x-ray CT image prior. A computer simulation study was conducted to compare the autoencoder CNN-derived feature representation with raw image patches for evaluation of kernel MLAA for GCT image reconstruction and dual-energy multimaterial decomposition. The results show that the autoencoder kernel MLAA method can achieve a significant image quality improvement for GCT and material decomposition as compared to the existing kernel MLAA algorithm. A weakness of the proposed method is its potential over-smoothness in a bone region, indicating the importance of further optimization in future work. The codes is available on https://github.com/SiqiLi1020/Autoencoder- Kernel-MLAA.