Signal Enhancement of Low Magnetic Field Magnetic Resonance Image Using a Conventional- and Cyclic-Generative Adversarial Network Models With Unpaired Image Sets.

Denis Yoo,Jing Cai,Eric Lee,Byung Jun Min,Yuni Annette Choi,Eun Ho Kim,C J Rah

doi:10.3389/fonc.2021.660284

Abstract

In this study, the signal enhancement ratio of low-field magnetic resonance (MR) images was investigated using a deep learning-based algorithm. Unpaired image sets (0.06 Tesla and 1.5 Tesla MR images for different patients) were used in this study following three steps workflow. In the first step, the deformable registration of a 1.5 Tesla MR image into a 0.06 Tesla MR image was performed to ensure that the shapes of the unpaired set matched. In the second step, a cyclic-generative adversarial network (GAN) was used to generate a synthetic MR image of the original 0.06 Tesla MR image based on the deformed or original 1.5 Tesla MR image. Finally, an enhanced 0.06 Tesla MR image could be generated using the conventional-GAN with the deformed or synthetic MR image. The results from the optimized flow and enhanced MR images showed significant signal enhancement of the anatomical view, especially in the nasal septum, inferior nasal choncha, nasopharyngeal fossa, and eye lens. The signal enhancement ratio, signal-to-noise ratio (SNR) and correlation factor between the original and enhanced MR images were analyzed for the evaluation of the image quality. A combined method using conventional- and cyclic-GANs is a promising approach for generating enhanced MR images from low-magnetic-field MR.

Highlights

Magnetic resonance imaging (MRI)-based contouring is a standard practice in radiotherapy [1,2,3,4]
In the 1st step, midpoint independent deformable registration was applied to the original 1.5 Tesla magnetic resonance (MR) image according to the 0.06 Tesla original MR image
This study investigated a deep learning-based three-step workflow to increase the signal intensity for MR images and determine feasibility of generating an enhanced MR image with an unpaired image set

Summary

Introduction

Magnetic resonance imaging (MRI)-based contouring is a standard practice in radiotherapy [1,2,3,4]. For proton therapy, the feasibility of magnetic resonance (MR)-only treatment planning has been investigated [17, 18]. In this MR-only RT workflow, the development of techniques for determining the electron density in MRI-derived substitute-computed tomography (sCT) images have been investigated, including atlas-based and deep learning-based methods [18,19,20,21,22]. For synthetic CT generation, Han’s generative adversarial network (GAN) model [20] or modified models [18, 21, 22] have generally been used with sets of twopaired images (an MR/CT image set for the same patient taken within one day) In this conventional GAN, the CT images are the ground truth, and the MR images are the input images

Objectives

Results

Discussion

Conclusion