Improving phase-based conductivity reconstruction by means of deep learning–based denoising of B1+ phase data for 3T MRI

Kyu Jin Jung, Stefano Mandija, Jun Hyeong Kim, Kanghyun Ryu, Soozy Jung, Chuanjiang Cui, Soo Yeon Kim, Mina Park, Cornelis A.T. van den Berg, Dong Hyun Kim

Research output: Contribution to journalArticlepeer-review

Abstract

Purpose: To denoise (Formula presented.) phase using a deep learning method for phase-based in vivo electrical conductivity reconstruction in a 3T MR system. Methods: For (Formula presented.) phase deep-learning denoising, a convolutional neural network (U-net) was chosen. Training was performed on data sets from 10 healthy volunteers. Input data were the real and imaginary components of single averaged spin-echo data (SNR = 45), which was used to approximate the (Formula presented.) phase. For label data, multiple signal-averaged spin-echo data (SNR = 128) were used. Testing was performed on in silico and in vivo data. Reconstructed conductivity maps were derived using phase-based conductivity reconstructions. Additionally, we investigated the usability of the network to various SNR levels, imaging contrasts, and anatomical sites (ie, T1, T2, and proton density–weighted brain images and proton density–weighted breast images. In addition, conductivity reconstructions from deep learning–based denoised data were compared with conventional image filters, which were used for data denoising in electrical properties tomography (ie, the Gaussian filtering and the Savitzky-Golay filtering). Results: The proposed deep learning–based denoising approach showed improvement for (Formula presented.) phase for both in silico and in vivo experiments with reduced quantitative error measures compared with other methods. Subsequently, this resulted in an improvement of reconstructed conductivity maps from the denoised (Formula presented.) phase with deep learning. Conclusion: The results suggest that the proposed approach can be used as an alternative preprocessing method to denoise (Formula presented.) maps for phase-based conductivity reconstruction without relying on image filters or signal averaging.

Original languageEnglish
Pages (from-to)2084-2094
Number of pages11
JournalMagnetic Resonance in Medicine
Volume86
Issue number4
DOIs
Publication statusPublished - 2021 Oct

Bibliographical note

Funding Information:
The National Research Foundation of Korea, funded by the Korea government (NRF-2019R1A2C1090635) and the Ministry of Science and ICT, Korea, under the Information Technology Research Center support program (IITP-2020-2020-0-01461), which is supervised by the Institute for Information & Communications Technology Planning & Evaluation, Korea

Publisher Copyright:
© 2021 International Society for Magnetic Resonance in Medicine

All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Improving phase-based conductivity reconstruction by means of deep learning–based denoising of B1+ phase data for 3T MRI'. Together they form a unique fingerprint.

Cite this