The purpose of this study was to evaluate the feasibility of using integrated slice-by-slice shimming (iShim) for improving the image quality of whole-body diffusion-weighted imaging (WBDWI) in patients with plasma disorder at 3 T by comparing to WBDWI using a conventional shimming (3-dimensional [3D] shim) adjustment of each body station.
Materials and Methods
After approval by the local institutional review board, 2 healthy volunteers and 29 suspected patients with plasma disorder participated in this cross-sectional study. All participants were scanned by the same WBDWI protocol with iShim and 3D shim, consecutively. Body region–dependent signal-to-noise ratio (SNR) between iShim and 3D shim was compared in 2 volunteer scans. Body region–dependent shimming parameters, image quality, the number of suspicious lesions, and the agreement of apparent diffusion coefficient values were compared in 29 suspected patients with plasma disorder. Signal integrity between images acquired at adjacent bed positions, in sagittal reformats, which may be impaired by susceptibility effects, was assessed using an image quality score between 1 (worst) and 4 (best). Spatial displacement of diffusion-weighted images was calculated using a 3D turbo spin-echo acquisition as a reference.
For the SNR comparison in 2 volunteers, the iShim technique yielded 8-fold and 15-fold SNR improvements in the neck region (P < 0.05), while keeping a comparable SNR (ratio, 1 ± 0.2) performance in other body regions (P > 0.05). In the patient group, the mean score of image quality for iShim WBDWI was 3.69, and 76% of the cases showed unimpaired whole-body signal integrity, while in 3D shim WBDWI images, the mean image quality score was 1.93, and only 7% of the cases showed unimpaired whole-body signal integrity. The spatial displacement of diffusion-weighted images was on average reduced from 7.21 mm (3D shim) to 3.89 mm by using the iShim technique.
Twenty-four of 72 suspicious lesions visible in iShim images of the neck region were missed on 3D shim images due to signal loss, while the number of focal lesions in the other body regions was comparable. With the 2 different shimming techniques, the agreement of apparent diffusion coefficient values of lesion and muscle in the head, thorax, abdomen, and pelvis regions was excellent (r > 0.75; P > 0.05), but there was a significant difference in the neck region (P < 0.05).
The iShim technique is an effective method to reduce the negative impact of susceptibility effects at 3 T on whole-body diffusion imaging, as supported by the apparent improvement in signal integrity and spatial alignment, as well as improved SNR in the neck region. Compared with the 3D shim technique, the iShim technique showed improved conspicuous lesion findings in the neck region.