Thin-slice Two-dimensional T2-weighted Imaging with Deep Learning-based Reconstruction: Improved Lesion Detection in the Brain of Patients with Multiple Sclerosis

Abstract

Purpose: Brain MRI with high spatial resolution allows for a more detailed delineation of multiple sclerosis (MS) lesions. The recently developed deep learning-based reconstruction (DLR) technique enables image denoising with sharp edges and reduced artifacts, which improves the image quality of thin-slice 2D MRI. We, therefore, assessed the diagnostic value of 1 mm-slice-thickness 2D T2-weighted imaging (T2WI) with DLR (1 mm T2WI with DLR) compared with conventional MRI for identifying MS lesions.Methods: Conventional MRI (5 mm T2WI, 2D and 3D fluid-attenuated inversion recovery) and 1 mm T2WI with DLR (imaging time: 7 minutes) were performed in 42 MS patients. For lesion detection, two neuroradiologists counted the MS lesions in two reading sessions (conventional MRI interpretation with 5 mm T2WI and MRI interpretations with 1 mm T2WI with DLR). The numbers of lesions per region category (cerebral hemisphere, basal ganglia, brain stem, cerebellar hemisphere) were then compared between the two reading sessions.Results: For the detection of MS lesions by 2 neuroradiologists, the total number of detected MS lesions was significantly higher for MRI interpretation with 1 mm T2WI with DLR than for conventional MRI interpretation with 5 mm T2WI (765 lesions vs. 870 lesions at radiologist A, < 0.05). In particular, of the 33 lesions in the brain stem, radiologist A detected 21 (63.6%) additional lesions by 1 mm T2WI with DLR. Conclusion:Using the DLR technique, whole-brain 1 mm T2WI can be performed in about 7 minutes, which is feasible for routine clinical practice. MRI with 1 mm T2WI with DLR enabled increased MS lesion detection, particularly in the brain stem.

Deep Learning-based Image Enhancement Techniques for Fast MRI in Neuroimaging

“…Similarly, Iwamura et al 32 have investigated a similar hypothesis for thin-slice brain imaging in 42 patients with multiple sclerosis (MS). In their study, the authors compared conventional 5 mm 2D T2WI with 1 mm 2D T2WI and applied a DL-based reconstruction software (AIR Recon DL) to the latter.…”

Improving Diagnostic Performance of MRI for Temporal Lobe Epilepsy With Deep Learning-Based Image Reconstruction in Patients With Suspected Focal Epilepsy

“…DLR learns to reconstruct images by recognizing patterns of low resolution and noise through training on previous data, thereby reconstructing only the ideal image [ 15 ]. Although previous studies have shown improved SNR and contrast-to-noise ratio (CNR) using DLR techniques [ 16 17 18 19 20 21 ], the utility of DLR for detecting abnormalities in the diagnosis of TLE has not been evaluated. We hypothesized that 1.5-mm slice-thickness oblique coronal 2D turbo spin echo T2-weighted imaging reconstructed with DLR (1.5-mm MRI + DLR) would be superior to routine 3-mm slice-thickness oblique coronal 2D turbo spin echo T2-weighted imaging (routine MRI) and 1.5-mm slice thickness MRI without DLR (1.5-mm MRI without DLR) in detecting focal lesions in patients with TLE.…”

Ultrafast Brain MRI at 3 T for MS: Evaluation of a 51-Second Deep-Learning-Enhanced T2-EPI-FLAIR Sequence

“…In recent years, significant advancements in machine learning have gained great attention in the field of medical imaging [15]. Applied to image reconstruction, these deep learning (DL) techniques provide an improved trade-off between speed, resolution and signal-to-noise ratio (SNR), and often enable significant reductions in scan times when combined with (highly) accelerated conventional techniques such as parallel imaging (PI) [16][17][18][19]. Alternative methods include compressed sensing (CS) [20][21][22][23], simultaneous multislice (SMS) imaging (also known as multiband imaging) [24][25][26][27][28], iterative denoising (ID) [29,30] and synthetic MRI [31][32][33][34][35][36][37][38].…”