Search

Background: Evaluation of prostate MRI relies on diffusion-weighted imaging (DWI), commonly distorted by susceptibility artifacts, thereby creating a need for approaches to correct such distortion.Purpose: To compare geometric distortion on prostate MRI between standard DWI and a geometric distortion correction method for DWI described as static distortion correction DWI (SDC DWI).Study Type: Retrospective case study.Population: Thirty patients (ages 31-81 years) undergoing prostate MRI.Sequence: Geometric distortions from echo planar imaging were corrected with the SDC DWI protocol, which first acquires a B0 -field map to estimate geometric distortions.Assessment: Contours of the prostate were placed on axial T2 -weighted imaging (T2 WI) as an anatomic standard. Pixel shifts and apparent diffusion coefficient (ADC) values were compared between prostate contours applied to the SDC DWI and standard DWI sequences. Detailed characterization of the impact of SDC DWI was performed in three representative patients.Statistical Tests: One-way analysis of variance (ANOVA) test, Spearman correlation test, and Bland-Altman plots were calculated.Results: There was significantly greater overlap of the SDC DWI prostate region of interest (ROI) with T2 WI than standard DWI with T2 WI (10.56 cm2 ± 3.14, P < 0.05). R2 of ADC values from standard DWI vs. SDC DWI in the 30 patients ranged from 0.02-0.94 (mean 0.60). A patient without susceptibility artifact demonstrated minimal pixel shift ranging from 0.6-1.3 mm and high correlation of ADC values (R2 = 0.89) between SDC DWI and standard DWI. A patient with rectal gas showed greater pixel shift (range: -2.5 to -0.5 mm) and less ADC value correlation (R2 = 0.69). A patient with a pelvic phlebolith adjacent to the prostate showed an even greater pixel shift (range: 10-16 mm) and decreased ADC correlation (R2 = 0.21).Data Conclusion: SDC DWI appears to correct for susceptibility-related pixel shifts in the prostate compared with standard DWI, which may have value for assessing prostate lesions obscured by geometric warping. Level of Evidence 4 Technical Efficacy Stage 1 J. Magn. Reson. Imaging 2019;50:1614-1619. [ABSTRACT FROM AUTHOR]

Purpose: To develop and test a deep learning approach named Convolutional Neural Network (CNN) for automated screening of T2 -weighted (T2 WI) liver acquisitions for nondiagnostic images, and compare this automated approach to evaluation by two radiologists.Materials and Methods: We evaluated 522 liver magnetic resonance imaging (MRI) exams performed at 1.5T and 3T at our institution between November 2014 and May 2016 for CNN training and validation. The CNN consisted of an input layer, convolutional layer, fully connected layer, and output layer. 351 T2 WI were anonymized for training. Each case was annotated with a label of being diagnostic or nondiagnostic for detecting lesions and assessing liver morphology. Another independently collected 171 cases were sequestered for a blind test. These 171 T2 WI were assessed independently by two radiologists and annotated as being diagnostic or nondiagnostic. These 171 T2 WI were presented to the CNN algorithm and image quality (IQ) output of the algorithm was compared to that of two radiologists.Results: There was concordance in IQ label between Reader 1 and CNN in 79% of cases and between Reader 2 and CNN in 73%. The sensitivity and the specificity of the CNN algorithm in identifying nondiagnostic IQ was 67% and 81% with respect to Reader 1 and 47% and 80% with respect to Reader 2. The negative predictive value of the algorithm for identifying nondiagnostic IQ was 94% and 86% (relative to Readers 1 and 2).Conclusion: We demonstrate a CNN algorithm that yields a high negative predictive value when screening for nondiagnostic T2 WI of the liver.Level Of Evidence: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;47:723-728. [ABSTRACT FROM AUTHOR]