Thin-Slice Prostate MRI Enabled by Deep Learning Image Reconstruction

Simple Summary Multi-parametric MRI (mpMRI) of the prostate is one emerging tool for early detection of clinically-significant cancer. Standard acquisition protocols provide a slice thickness of 3 mm in T2-weighted TSE imaging. However, thin-slice imaging might be superior for the assessment of the prostate parenchyma. The main disadvantage of thin-slice imaging is related to prolongation of acquisition time. In this study, similar acquisition times could be enabled by deep learning image reconstruction of thin-slice imaging as compared to standard 3 mm T2 imaging. The results demonstrate superior image quality and diagnostic confidence in thin-slice imaging. Objectives: Thin-slice prostate MRI might be beneficial for prostate cancer diagnostics. However, prolongation of acquisition time is a major drawback of thin-slice imaging. Therefore, the purpose of this study was to investigate the impact of a thin-slice deep learning accelerated T2-weighted (w) TSE imaging sequence (T2(DLR)) of the prostate as compared to conventional T2w TSE imaging (T2(S)). Materials and Methods: Thirty patients were included in this prospective study at one university center after obtaining written informed consent. T2(S) (3 mm slice thickness) was acquired first in three orthogonal planes followed by thin-slice T2(DLR) (2 mm slice thickness) in axial plane. Acquisition time of axial conventional T2(S) was 4:12 min compared to 4:37 min for T2(DLR). Imaging datasets were evaluated by two radiologists using a Likert-scale ranging from 1-4, with 4 being the best regarding the following parameters: sharpness, lesion detectability, artifacts, overall image quality, and diagnostic confidence. Furthermore, preference of T2(S) versus T2(DLR) was evaluated. Results: The mean patient age was 68 +/- 8 years. Sharpness of images and lesion detectability were rated better in T2(DLR) with a median of 4 versus a median of 3 in T2(S) (p < 0.001 for both readers). Image noise was evaluated to be significantly worse in T2(DLR) as compared to T2(S) (p < 0.001 and p = 0.021, respectively). Overall image quality was also evaluated to be superior in T2(DLR) versus T2(S) with a median of 4 versus 3 (p < 0.001 for both readers). Both readers chose T2(DLR) in 29 cases as their preference. Conclusions: Thin-slice T2(DLR) of the prostate provides a significant improvement of image quality without significant prolongation of acquisition time.