Deep Learning Reconstruction of Diffusion-weighted MRI Improves Image Quality for Prostatic Imaging.

Abstract

Background Deep learning reconstruction (DLR) may improve image quality. However, its impact on diffusion-weighted imaging (DWI) of the prostate has yet to be assessed. Purpose To determine whether DLR can improve image quality of diffusion-weighted MRI at <i>b</i> values ranging from 1000 sec/mm² to 5000 sec/mm² in patients with prostate cancer. Materials and Methods In this retrospective study, images of the prostate obtained at DWI with a <i>b</i> value of 0 sec/mm², DWI with a <i>b</i> value of 1000 sec/mm² (DWI₁₀₀₀), DWI with a <i>b</i> value of 3000 sec/mm² (DWI₃₀₀₀), and DWI with a <i>b</i> value of 5000 sec/mm² (DWI₅₀₀₀) from consecutive patients with biopsy-proven cancer from January to June 2020 were reconstructed with and without DLR. Image quality was assessed using signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) from region-of-interest analysis and qualitatively assessed using a five-point visual scoring system (1 [very poor] to 5 [excellent]) for each high-<i>b</i>-value DWI sequence with and without DLR. The SNR, CNR, and visual score for DWI with and without DLR were compared with the paired <i>t</i> test and the Wilcoxon signed rank test with Bonferroni correction, respectively. Apparent diffusion coefficients (ADCs) from DWI with and without DLR were also compared with the paired <i>t</i> test with Bonferroni correction. Results A total of 60 patients (mean age, 67 years; age range, 49-79 years) were analyzed. DWI with DLR showed significantly higher SNRs and CNRs than DWI without DLR (<i>P</i> < .001); for example, with DWI₁₀₀₀ the mean SNR was 38.7 ± 0.6 versus 17.8 ± 0.6, respectively (<i>P</i> < .001), and the mean CNR was 18.4 ± 5.6 versus 7.4 ± 5.6, respectively (<i>P</i> < .001). DWI with DLR also demonstrated higher qualitative image quality than DWI without DLR (mean score: 4.8 ± 0.4 vs 4.0 ± 0.7, respectively, with DWI₁₀₀₀ [<i>P</i> = .001], 3.8 ± 0.7 vs 3.0 ± 0.8 with DWI₃₀₀₀ [<i>P</i> = .002], and 3.1 ± 0.8 vs 2.0 ± 0.9 with DWI₅₀₀₀ [<i>P</i> < .001]). ADCs derived with and without DLR did not differ substantially (<i>P</i> > .99). Conclusion Deep learning reconstruction improves the image quality of diffusion-weighted MRI scans of prostate cancer with no impact on apparent diffusion coefficient quantitation with a 3.0-T MRI system. © RSNA, 2022 <i>Online supplemental material is available for this article</i>. See also the editorial by Turkbey in this issue.