Literature DB >> 32542779

Combined Denoising and Suppression of Transient Artifacts in Arterial Spin Labeling MRI Using Deep Learning.

Patrick W Hales1, Josef Pfeuffer2, Chris A Clark1.   

Abstract

BACKGROUND: Arterial spin labeling (ASL) is a useful tool for measuring cerebral blood flow (CBF). However, due to the low signal-to-noise ratio (SNR) of the technique, multiple repetitions are required, which results in prolonged scan times and increased susceptibility to artifacts.
PURPOSE: To develop a deep-learning-based algorithm for simultaneous denoising and suppression of transient artifacts in ASL images. STUDY TYPE: Retrospective.
SUBJECTS: 131 pediatric neuro-oncology patients for model training and 11 healthy adult subjects for model evaluation. FIELD STRENGTH/SEQUENCE: 3T / pseudo-continuous and pulsed ASL with 3D gradient-and-spin-echo readout. ASSESSMENT: A denoising autoencoder (DAE) model was designed with stacked encoding/decoding convolutional layers. Reference standard images were generated by averaging 10 pairwise ASL subtraction images. The model was trained to produce perfusion images of a similar quality using a single subtraction image. Performance was compared against Gaussian and non-local means (NLM) filters. Evaluation metrics included SNR, peak SNR (PSNR), and structural similarity index (SSIM) of the CBF images, compared to the reference standard. STATISTICAL TESTS: One-way analysis of variance (ANOVA) tests for group comparisons.
RESULTS: The DAE model was the only model to produce a significant increase in SNR compared to the raw images (P < 0.05), providing an average SNR gain of 62%. The DAE model was also effective at suppressing transient artifacts, and was the only model to show a significant improvement in accuracy in the generated CBF images, as assessed using PSNR values (P < 0.05). In addition, using data from multiple inflow time acquisitions, the DAE images produced the best fit to the Buxton kinetic model, offering a 75% reduction in the fitting error compared to the raw images. DATA
CONCLUSION: Deep-learning-based algorithms provide superior accuracy when denoising ASL images, due to their ability to simultaneously increase SNR and suppress artifactual signals in raw ASL images. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 1.
© 2020 The Authors. Journal of Magnetic Resonance Imaging published by Wiley Periodicals, Inc. on behalf of International Society for Magnetic Resonance in Medicine.

Entities:  

Keywords:  ASL; CNN; arterial spin labeling; autoencoder; deep learning; denoising

Mesh:

Substances:

Year:  2020        PMID: 32542779     DOI: 10.1002/jmri.27255

Source DB:  PubMed          Journal:  J Magn Reson Imaging        ISSN: 1053-1807            Impact factor:   4.813


  3 in total

1.  Suppression of artifact-generating echoes in cine DENSE using deep learning.

Authors:  Mohamad Abdi; Xue Feng; Changyu Sun; Kenneth C Bilchick; Craig H Meyer; Frederick H Epstein
Journal:  Magn Reson Med       Date:  2021-05-22       Impact factor: 3.737

Review 2.  Recent Technical Developments in ASL: A Review of the State of the Art.

Authors:  Luis Hernandez-Garcia; Verónica Aramendía-Vidaurreta; Divya S Bolar; Weiying Dai; Maria A Fernández-Seara; Jia Guo; Ananth J Madhuranthakam; Henk Mutsaerts; Jan Petr; Qin Qin; Jonas Schollenberger; Yuriko Suzuki; Manuel Taso; David L Thomas; Matthias J P van Osch; Joseph Woods; Moss Y Zhao; Lirong Yan; Ze Wang; Li Zhao; Thomas W Okell
Journal:  Magn Reson Med       Date:  2022-08-19       Impact factor: 3.737

3.  Applicability of deep learning-based reconstruction trained by brain and knee 3T MRI to lumbar 1.5T MRI.

Authors:  Nobuo Kashiwagi; Hisashi Tanaka; Yuichi Yamashita; Hiroto Takahashi; Yoshimori Kassai; Masahiro Fujiwara; Noriyuki Tomiyama
Journal:  Acta Radiol Open       Date:  2021-06-18
  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.