Two-Minute Quantitative Susceptibility Mapping From Three-Dimensional Echo-Planar Imaging

Abstract

Objectives The aim of this study was to assess the accuracy, reliability, and cerebral microbleed (CMB) detection performance of 2-minute quantitative susceptibility mapping (QSM) from 3-dimensional echo-planar imaging (3D-EPI). Materials and Methods Gadolinium phantom study was conducted using 3D-EPI, single–echo time (TE), and multi-TE gradient-recalled echo (GRE) sequences on two 3-T magnetic resonance (MR) scanners to assess the accuracy between measured and theoretical susceptibility values. The institutional review board approved this prospective study, and 40 healthy volunteers were enrolled with written consent between April 2018 and October 2019. Each underwent 3D-EPI, single-TE, and multi-TE GRE sequences consecutively on one 3-T MR scanner, and QSMs were calculated to assess the reliability of 3D-EPI QSM. Intraclass correlation coefficient (ICC), linear regression, and Bland-Altman plots were calculated. Patients with CMB who underwent both 3D-EPI and GRE QSM scans were retrospectively enrolled. Two radiologists evaluated images independently, and Cohen κ coefficients were calculated to compare CMB detection performance. Results Phantom study showed excellent validity of 3D-EPI QSM on both MR scanners: Skyra, R 2 = 0.996, P < 0.001, ICC = 0.997, mean difference, −2 ppb (95% confidence interval [CI], −45 to 40 ppb); Prisma, R 2 = 0.992, P < 0.001, ICC = 0.988, mean difference, 15 ppb (95% CI, −67 to 97 ppb). A human study of 40 healthy volunteers (59 ± 13 years, 25 women) showed excellent reliability with 3D-EPI QSM for both single-TE and multi-TE GRE (R 2 = 0.981, P < 0.001, ICC = 0.988; R 2 = 0.983, P < 0.001, ICC = 0.990, respectively), supported by a Bland-Altman mean difference of 4 ppb (95% CI, −15 to 23 ppb) for single-TE GRE and 3 ppb (95% CI, −15 to 20 ppb) for multi-TE GRE. The CMB detection performance evaluation from 38 patients (51 ± 20 years, 20 women) showed almost perfect agreement between 3D-EPI and GRE QSM for both raters (κ = 0.923 and 0.942, P < 0.001). Conclusions Faster QSM from 3D-EPI demonstrated excellent accuracy, reliability, and CMB detection performance.

Iron Deposition Characteristics of Deep Gray Matter in Elderly Individuals in the Community Revealed by Quantitative Susceptibility Mapping and Multiple Factor Analysis

“…In the cross-sectional analysis, age-related iron deposition was found in the GN, Put, CN, and DN (Xu et al, 2008;Li et al, 2014;Gong et al, 2015). Based on the knowledge that QSM is reproducible across scanner makers, models, field strengths, and sites and could be used in clinical investigations both longitudinally and across centers (Hinoda et al, 2015;Wang et al, 2017;Wicaksono et al, 2020), the MR measures of QSM were accurate and consistent across repeated measurements and between platforms (Hobson et al, 2020). We performed a longitudinal analysis of age-related iron deposition despite different scanners in the two examinations.…”

Increased Brain Iron Deposition in the Putamen in Patients with Type 2 Diabetes Mellitus Detected by Quantitative Susceptibility Mapping

“…Meanwhile, microbleeds change local magnetic susceptibility because of the pathologic iron accumulation [ 43 ]. QSM demonstrated excellent accuracy and reliability in detecting microbleeds [ 44 ]. Our findings also supported that microvascular hemorrhage may be responsible for the hyperintensity on T1-weighted images.…”

QSM Reconstruction Challenge 2.0: a realistic in silico head phantom for MRI data simulation and evaluation of susceptibility mapping procedures

“…Spatial distortions associated with different readouts can be obtained in computationally efficient manner by using our provided phantom data in, for example, freely available software packages such as JEMRIS, http://www.jemris.org/ (77). Wave Caipi (78) and 3D EPI (79,80) have been used for QSM, but not much research has been done to quantify the impact of their bluring or distortions on the performance of neither the background field removal nor the susceptibility maps. Respiration artifacts would simply require a library of respiration fields over time as acquired with field cameras (81).…”