Super‐resolution head and neck MRA using deep machine learning

Abstract

To probe the feasibility of deep learning-based super-resolution (SR) reconstruction applied to nonenhanced MR angiography (MRA) of the head and neck. Methods: High-resolution 3D thin-slab stack-of-stars quiescent interval sliceselective (QISS) MRA of the head and neck was obtained in eight subjects (seven healthy volunteers, one patient) at 3T. The spatial resolution of high-resolution ground-truth MRA data in the slice-encoding direction was reduced by factors of 2 to 6. Four deep neural network (DNN) SR reconstructions were applied, with two based on U-Net architectures (2D and 3D) and two (2D and 3D) consisting of serial convolutions with a residual connection. SR images were compared to ground-truth high-resolution data using Dice similarity coefficient (DSC), structural similarity index measure (SSIM), arterial diameter, and arterial sharpness measurements. Image review of the optimal DNN SR reconstruction was done by two experienced neuroradiologists. Results: DNN SR of up to twofold and fourfold lower-resolution (LR) input volumes provided images that resembled those of the original high-resolution groundtruth volumes for intracranial and extracranial arterial segments, and improved DSC, SSIM, arterial diameters, and arterial sharpness relative to LR volumes (P < .001). The 3D DNN SR outperformed 2D DNN SR reconstruction. According to two neuroradiologists, 3D DNN SR reconstruction consistently improved image quality with respect to LR input volumes (P < .001). Conclusion: DNN-based SR reconstruction of 3D head and neck QISS MRA offers the potential for up to fourfold reduction in acquisition time for neck vessels without the need to commensurately sacrifice spatial resolution.

Analysis of a Deep Learning-Based Superresolution Algorithm Tailored to Partial Fourier Gradient Echo Sequences of the Abdomen at 1.5 T

“…19 The successful application of superresolution was previously demonstrated in head and neck and knee imaging as well as in precontrast and postcontrast abdominal MRI. 19,31,32 The novelty of the presented algorithm in our study consists especially of the partial omission of the acquired data with slight reduction of acquisition time and, therefore, reduction of breath-holds including possible mitigation of motion of the intestine or diaphragm leading to an increase in sharpness. Breathing artifacts are still one of the most significant impairments in GRE imaging of the upper abdomen.…”

Deep Learning Applications in Magnetic Resonance Imaging: Has the Future Become Present?

“…In contrast to simple denoising algorithms, super-resolution aims to increase the spatial resolution via DL-based post-processing [ 40 , 41 , 42 ]. This concept was successfully implemented in head and neck imaging, as well as abdominal and cardiac imaging [ 40 , 43 , 44 ]. Especially fast sequences, such as gradient echo (GRE) imaging, benefit from these implementations, due to their relatively low signal-to-noise ratios.…”

Deep Learning and Its Application to Function Approximation for MR in Medicine: An Overview

“…It was shown that a DNN was efficient as a super-resolution method for musculoskeletal images 97 and MR angiography (MRA). 98…”