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

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 devel...

Full description

Saved in:
Bibliographic Details
Published in:Journal of magnetic resonance imaging Vol. 52; no. 5; pp. 1413 - 1426
Main Authors: Hales, Patrick W., Pfeuffer, Josef, Clark, Chris
Format: Journal Article
Language:English
Published: Hoboken, USA John Wiley & Sons, Inc 01.11.2020
Wiley Subscription Services, Inc
Subjects:
Adult
Algorithms
arterial spin labeling
ASL
autoencoder
Blood flow
Brain - diagnostic imaging
Cerebral blood flow
Cerebrovascular Circulation
Child
CNN
Deep Learning
denoising
Encoding-Decoding
Field strength
Humans
Image acquisition
Image quality
Labeling
Magnetic Resonance Imaging
Medical imaging
Noise reduction
Oncology
Perfusion
Retrospective Studies
Simultaneous discrimination learning
Spin labeling
Spin Labels
Statistical analysis
Statistical tests
Variance analysis
denoising
deep learning
CNN
ASL
autoencoder
arterial spin labeling
ISSN:1053-1807, 1522-2586, 1522-2586
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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
Bibliography:Contract grant sponsor: Children with Cancer UK; Contract grant number CwCUK‐15‐203.
Level of EvidenceTechnical Efficacy Stage
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1053-1807
1522-2586
1522-2586
DOI:10.1002/jmri.27255