Deep learning–based acceleration of Compressed Sense MR imaging of the ankle

Abstract

Objectives To evaluate a compressed sensing artificial intelligence framework (CSAI) to accelerate MRI acquisition of the ankle. Methods Thirty patients were scanned at 3T. Axial T2-w, coronal T1-w, and coronal/sagittal intermediate-w scans with fat saturation were acquired using compressed sensing only (12:44 min, CS), CSAI with an acceleration factor of 4.6–5.3 (6:45 min, CSAI2x), and CSAI with an acceleration factor of 6.9–7.7 (4:46 min, CSAI3x). Moreover, a high-resolution axial T2-w scan was obtained using CSAI with a similar scan duration compared to CS. Depiction and presence of abnormalities were graded. Signal-to-noise and contrast-to-noise were calculated. Wilcoxon signed-rank test and Cohen’s kappa were used to compare CSAI with CS sequences. Results The correlation was perfect between CS and CSAI2x (κ = 1.0) and excellent for CS and CSAI3x (κ = 0.86–1.0). No significant differences were found for the depiction of structures between CS and CSAI2x and the same abnormalities were detected in both protocols. For CSAI3x the depiction was graded lower (p ≤ 0.001), though most abnormalities were also detected. For CSAI2x contrast-to-noise fluid/muscle was higher compared to CS (p ≤ 0.05), while no differences were found for other tissues. Signal-to-noise and contrast-to-noise were higher for CSAI3x compared to CS (p ≤ 0.05). The high - resolution axial T2-w sequence specifically improved the depiction of tendons and the tibial nerve (p ≤ 0.005). Conclusions Acquisition times can be reduced by 47% using CSAI compared to CS without decreasing diagnostic image quality. Reducing acquisition times by 63% is feasible but should be reserved for specific patients. The depiction of specific structures is improved using a high-resolution axial T2-w CSAI scan. Key Points • Prospective study showed that CSAI enables reduction in acquisition times by 47% without decreasing diagnostic image quality. • Reducing acquisition times by 63% still produces images with an acceptable diagnostic accuracy but should be reserved for specific patients. • CSAI may be implemented to scan at a higher resolution compared to standard CS images without increasing acquisition times.

Knowledge‐driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un‐supervised learning

“…49,50 Most are based on the image domain SENSE model for coil combination or for building the DC term. [51][52][53] Some methods attempt to utilize the intrinsic relationship between coils using neural networks, without explicitly estimating the sensitivity maps of the coils. [54][55][56][57][58][59] Other methods attempt to simultaneously estimate coil sensitivity maps and perform image reconstruction.…”

Reconstruction of 3D knee MRI using deep learning and compressed sensing: a validation study on healthy volunteers

“…Notably, Fervers et al observed significantly improved subjective image quality for 3D T2-weighted images of the lumbar spine when employing CS-AI, as opposed to CS-based reconstruction [ 20 ]. Moreover, studies focusing on MRI of the ankle and prostate have reported enhanced objective and subjective image quality for sequences reconstructed with CS-AI [ 21 , 22 ]. While CS-AI yields images with superior subjective image quality in comparison to CS, it is important to consider that excessively high acceleration levels may compromise diagnostic quality.…”

“…Third, it is important to note that our study did not incorporate objective measures of image quality. Previous research by Foreman et al has discussed that traditional objective quality measures, such as signal-to-noise ratio, may not accurately reflect the quality of accelerated images generated through DL techniques [ 21 ]. Interestingly, Foreman et al [ 21 ] observed an increase in signal-to-noise ratio with higher acceleration levels, despite a decrease in subjective image quality.…”

“…Previous research by Foreman et al has discussed that traditional objective quality measures, such as signal-to-noise ratio, may not accurately reflect the quality of accelerated images generated through DL techniques [ 21 ]. Interestingly, Foreman et al [ 21 ] observed an increase in signal-to-noise ratio with higher acceleration levels, despite a decrease in subjective image quality. This discrepancy could be attributed to the inherent noise reduction capabilities of DL-based reconstructions, rendering them less noisy compared to conventional reconstructions.…”

Deep-learning-based reconstruction of T2-weighted magnetic resonance imaging of the prostate accelerated by compressed sensing provides improved image quality at half the acquisition time

“…This CNN is applied prior coil channel combination, removing the noise from the images to obtain good image quality from accelerated acquisitions ( 29 ). The Adaptive-CS-Network employed in this work was pre-trained on 740,000 sparsifying MR images using both 1.5T and 3T images of various anatomies and contrasts ( 25 , 33 ). The prototype was adapted and optimized to run on up-to-date standard reconstruction hardware available in our 3T MR scanner ( 31 ).…”