Correction of Motion Artifacts Using a Multiscale Fully Convolutional Neural Network
- Resource Type
- Authors
- Tom Brosch; Christopher S. Hall; Nathan M. Cross; Jalal B. Andre; Axel Saalbach; Karsten Sommer
- Source
- AJNR Am J Neuroradiol
- Subject
- Male
Mean squared error
Image quality
Pipeline (computing)
ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION
Neuroimaging
Convolutional neural network
Motion (physics)
030218 nuclear medicine & medical imaging
Reduction (complexity)
Motion
03 medical and health sciences
0302 clinical medicine
Image Processing, Computer-Assisted
Humans
Medicine
Radiology, Nuclear Medicine and imaging
Retrospective Studies
ComputingMethodologies_COMPUTERGRAPHICS
Artifact (error)
business.industry
Adult Brain
Brain
Pattern recognition
Magnetic Resonance Imaging
Neural Networks, Computer
Neurology (clinical)
Artificial intelligence
Artifacts
business
030217 neurology & neurosurgery
Test data
- Language
- ISSN
- 1936-959X
0195-6108
BACKGROUND AND PURPOSE: Motion artifacts are a frequent source of image degradation in the clinical application of MR imaging (MRI). Here we implement and validate an MRI motion-artifact correction method using a multiscale fully convolutional neural network. MATERIALS AND METHODS: The network was trained to identify motion artifacts in axial T2-weighted spin-echo images of the brain. Using an extensive data augmentation scheme and a motion artifact simulation pipeline, we created a synthetic training dataset of 93,600 images based on only 16 artifact-free clinical MRI cases. A blinded reader study using a unique test dataset of 28 additional clinical MRI cases with real patient motion was conducted to evaluate the performance of the network. RESULTS: Application of the network resulted in notably improved image quality without the loss of morphologic information. For synthetic test data, the average reduction in mean squared error was 41.84%. The blinded reader study on the real-world test data resulted in significant reduction in mean artifact scores across all cases (P