4617
Multiband SENSE accelerated diffusion weighted imaging of the abdomen with CAIPIRINHA: Preliminary study on clinical applicability1Philips Healthcare, Gainesville, FL, United States, 2Radiology, University of Washington, Seattle, WA, United States, 3Radiology, Seattle Cancer Care Alliance, Seattle, WA, United States
Synopsis
DWI is rapidly becoming a modality of choice to detect, characterize and monitor malignant lesions. Although diffusion imaging has benefited greatly from multiband imaging in the brain, investigation on abdomen DWI using MB has been limited. The purpose of this study was to examine the feasibility of MB-SENSE for abdominal DWI. The study demonstrated that MB-SENSE can be used to accelerate abdominal DWI with drastically reduced acquisition time (~50%) without having a significant impact on image quality. Though, refinement of the MB-SENSE free-breathing single-shot EPI sequence is necessary to ascertain the clinical value of MB-SENSE in DWI for the abdomen.
Introduction
Early detection and diagnosis of hepatic tumor is an important step in clinical work. Diffusion weighted imaging (DWI) is rapidly becoming a modality of choice to detect, characterize and monitor malignant lesions1. For a successful multi b-value DWI scan, increased number of signal average is usually necessary for higher b values, resulting in lengthened scan time in a clinical setting. Multiband (MB) imaging2 has been introduced as an emerging fast acquisition technique by simultaneously exciting multiple slices. Diffusion imaging has benefited greatly from MB-SENSE mainly to study neurologic disorders in the cerebral tissues. However, investigation on abdomen DWI has been limited3,4. The purpose of this study was to examine the feasibility of MB-SENSE acceleration for abdominal DWI. The image quality of MB-SENSE DWI was evaluated and compared to a conventional DWI sequence.Methods
This study was approved by the institutional review board. 21 patients (mean age 56, range 24-80 years; 9 females) were enrolled and underwent MRI examinations for abdomen cancer screening. MRI acquisitions were performed on a 1.5T clinical scanner (Ingenia; Philips Healthcare, Best, Netherlands) using a 32-ch phased array torso coil. 36-44 axial slices of abdomen were acquired using a free-breathing single-shot EPI sequence with the following parameters: MB-SENSE factor of 2, FOV/2 CAIPIRINHA2 shift, TR/TE=625-939/66-70ms, echo train length=55, FOV=360x315mm2, voxel size=2.8x2.8x6mm3, in-plane SENSE factor of 2, SPIR fat suppression, b values of 0 and 800s/mm2, and the total acquisition time of 1:21-2:01 min. A comparable conventional DWI EPI sequence was served as the reference standard: TR/TE=1300-1800/63-67ms and the acquisition time was 2:42-3:50min, with otherwise identical parameters.
In addition to indicating the presence or absence of lesions, image quality of the DWI trace images was also assessed by an experienced radiologist for the following aspects: overall image quality, SNR rating, artifact and lesion sharpness, with a highest score indicating the most desirable exam. The total scores were compared between MB-SENSE accelerated and reference standard DWI using post-hoc paired t-test. A p-value less than 0.05 was considered statistically significant.
Results
Example abdomen diffusion-weighted images from the conventional (a, c) and MB-SENSE accelerated (b, d) EPI are shown in Figure 1, for a 65-yo female patient. Similar image appearance and anatomical details are presented in MB-SENSE DWI scans compared to reference standard, with a ~50% reduction in scan time. Figure 2 shows the example trace-weighted images acquired at b=800 s/mm2 using standard DWI (a, c) and MB-SENSE accelerated EPI sequences (b, d) for 38-yo male and 80-yo male patients. The same lesions were identified by both methods. Scoring results on the image quality was summarized in Table 1. MB-SENSE and standard reference showed comparable overall image quality being only minimally limited to optimal for diagnosis, and little to no significant artifacts. The statistical analysis showed no significant differences in overall image quality (p=0.082) and artifacts (p=0.096). Though, SNR rating and lesion sharpness showed significant lower scorings for MB-SENSE DWI when compared to conventional DWI. The drop in SNR is somewhat expected due to drastically reduced TR and longer TE in MB-SENSE acquisitions, which could also explain the degradation in lesion sharpness in certain cases.
Total of nine patients were identified with presence of lesions by both MB-SENSE and conventional DWI EPI. Among these patients, three had single lesions that were detected by both sequences. Four out of six patients who had multiple lesions were also identified by both methods (lesion numbers ranges 4-30). For the two remaining patients, MB-SENSE has been found to be missing on lesions less than 5 mm in size. Otherwise, lesion detectability for lesions greater than 5mm in size are identical between MB-SENSE and conventional DWI.
Discussion and Conclusion
The study demonstrated that MB-SENSE can be used to accelerate abdominal DWI with drastically reduced acquisition time (~50%) without having a significant impact on image quality. Although we were able to achieve comparable overall image quality and artifact for the two methods, statistically differences remains in SNR rating and lesion sharpness. Our hypothesis for the SNR drop is due to the reduction in TR and lengthened TE. We validated this by experimenting with a longer TR: 2.2 s for MB-SENSE DWI vs. 4.5 s for reference standard DWI, as illustrated in Figure 3, for a 52-yo male patient. Improved image quality and SNR rating can be observed in the trace-weighted images. MB-SENESE DWI with longer TR thus holds promise for improved SNR and detectability of smaller lesions. Consequently, refinement of the MB-SENSE accelerated free-breathing single-shot EPI sequence is necessary to ascertain the clinical value of MB-SENSE in DWI for the abdomen.Acknowledgements
The authors would like to acknowledge Dr. Jeffery Maki for the helpful discussion on the study set up.References
1. Lima M, Le Bihan D. Clinical intravoxel incoherent motion and diffusion MR imaging: past, present and future. Radiology. 2016; 278(1):13-32.
2. Setsompop K, Gagoski BA, Polimeni JR, et al. Blipped-controlled aliasing in parallel imaging for simultaneous multislice echo planar imaging with reduced g-factor penalty. Magn Reson Med. 2012; 67, 1210-1224.
3. Obele CC, Glielmi C, Ream J, et al. Simultaneous multislice accelerated free-breathing diffusion-weighted imaging of the liver at 3T. Abdominal Imaging. 2015; 40: 2323-2330.
4. Kenkel D, Wurnig MC, Filli L, et al. Whole-body diffusion imaging applying simultaneous multi-slice excitation. Rofo. 2016; 188(4): 381-388.
Figures
