Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Reproducibility and feasibility of optic nerve diffusion MRI techniques: single-shot echo-planar imaging (EPI), readout-segmented EPI, and reduced field-of-view diffusion-weighted imaging

Reproducibility and feasibility of optic nerve diffusion MRI techniques: single-shot echo-planar... Background: Diffusion-weighted imaging (DWI) is an essential technique for optic nerve diseases. However, the image quality of optic nerve DWI is decreased by the distortions and artifacts associated with conventional tech- niques. In order to establish this method as a critical tool in optic nerve diseases, reproducibility and feasibility of new technical and conventional approaches of DWI need to be systematically investigated. Methods: DWIs were acquired using ss-EPI, readout-segmented EPI (rs-EPI) DWI, and reduced field-of-view (rFOV ) DWI. 26 volunteers (mean age 31.2 years) underwent repeated MRI examinations in order to assess scan–rescan reproducibility and accuracy. The apparent diffusion coefficient (ADC) values (three ROIs were measured on each side) were determined to evaluate the reproducibility of each sequence and the differences between the three techniques. To quantify the geometric distortion artifacts, the length of optic nerve and the maximum angle of optic nerve were defined and compared to T2-weighted imaging. In addition, two readers evaluated four different aspects of image quality on 5-point Likert scales. Results: rs-EPI DWI (ICCs: 0.916, 0.797 and 0.781) and rFOV DWI (ICCs: 0.850, 0.595 and 0.750) showed higher repro- ducibility (ICCs: ROI, ROI and ROI ) of mean ADC value in all three ROIs than ss-EPI DWI (ICCs: 0.810, 0.442 and 0.379). 1 2 3 The quantitative analysis of geometric distortion yielded a higher agreement of both rs-EPI DWI and rFOV DWI with T2-weighted imaging than ss-EPI. rs-EPI DWI (2.38 ± 0.90) and rFOV DWI (2.46 ± 0.58) were superior to ss-EPI DWI (1.58 ± 0.64) with respect to overall image quality and other aspects of image quality, each with P < 0.05. The mean ADC values of rFOV DWI were significantly lower than those of rs-EPI DWI and ss-EPI DWI in all three ROIs (P < 0.001). Conclusions: Both rs-EPI DWI and rFOV-EPI DWI are suitable techniques for the assessment of diffusion restriction and provide significantly improved image quality compared with ss-EPI DWI. For methods using the same acquisition *Correspondence: jintianzzw@163.com Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, No. 1, Youyi Road, Yuzhong District, Chongqing, China Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhou et al. BMC Medical Imaging (2022) 22:96 Page 2 of 11 time, rFOV DWI is superior to ss-EPI DWI, while rs-EPI showed an overall superiority, although this technique took 47% longer to perform. Keywords: Optic nerves, Diffusion-weighted imaging, Zoomed EPI, RESOLVE MRI protocol Background MRI was performed with a 3T MR scanner (MAG- Optic nerve diseases cause severe visual disturbances NETOM Skyra, Siemens Healthcare, Erlangen, Ger- that, currently, cannot be diagnostically confirmed on many) using a 20-channel head matrix coil, and gradients ocular examination [1]. Orbital magnetic resonance with a peak amplitude of 45 mT/m and slew rate of imaging (MRI) is a widely utilized method to analyze 200  T/m/s. Imaging sequences included T2w imaging, various optic diseases such as optical neuritis. The wide and DWIs were acquired using ss-EPI, high-resolution application of orbital MRI in clinical also highly val- rs-EPI (readout segmentation of long variable echo ued functional MR technology, particularly diffusion- trains, RESOLVE , Siemens Healthcare), and a proto- weighted imaging (DWI) [2–5]. Single-shot echo-planar type rFOV-EPI (syngo ZOOMit , Siemens Healthcare). imaging (ss-EPI) is a conventional DW technique for The imaging protocol was repeated on each of the sub - rapid assessment of diffusion restriction and was com - jects at an interval of 20–40 min. Volunteers were placed monly used in clinical evaluation and differential diagno - in a comfortable supine position, and a sponge was posi- sis of optic neuritis [6, 7]. But in clinical circumstances, tioned to limit head movement. ss-EPI DWI of optic nerve suffers from partial-volume B0 inhomogeneity can result in a variety of artifacts, effects, magnetic susceptibility artifacts, chemical shift including geometric distortion, image blurring, especially artifacts, and low image quality on account of the small in EPI sequences [20]. For a fair comparison, the standard dimension of the nerves and low resolution [8]. Recent B0 shimming was used in all the sequences; The upside studies have indicated that reduced field-of-view (rFOV) of utilizing modeled shim fields is that there are no issues DWI provides outstanding improvements in image with different FOVs or different resolutions [20]. The quality of the spinal cord [9, 10], pancreas [11], pros- bandwidth was adjusted to the optimal value to make tate [12, 13], breast [14], thyroid nodules [15], and optic the echo spacing close to the minimum value to reduce nerve [16]. Literature has demonstrated that rFOV DWI image distortion in all three DWI sequences. Addition- showed an improvement in subjective image quality for ally, GRAPPA (GeneRalized Autocalibrating Partially optic nerve in the intraorbital segment compared with Parallel Acquisitions) was applied in each of the three ss-EPI [17, 18]. Moreover, Readout-segmented echo- DWI sequences to reduce susceptibility changes at tissue planar imaging (rs-EPI) is a promising technique that has interfaces [21]. already been reported to increase image quality in orbital The rFOV DWI sequence used the ‘Excitation Model’ imaging [19]. Therefore, the present study focused on of ZOOMit, replacing the traditional pulse with slice comparing the reproducibility and feasibility of current selective excitation. Different from conventional zoomed state-of-the-art diffusion techniques for their application technology, the advanced non-parallel transmission in optic nerve analyses. (non-PTX) zoomed-DWI was performed with a rota- tion of the field of excitation (8 deg.) to reduce distortion and echo time and correspondingly improve image qual- Methods ity despite B0 inhomogeneities [22]. In addition, averag- Subjects ing (10 averages) was applied to improve the evaluation From December 2019 to April 2020, 33 healthy volun- accuracy of the ADC values. Moreover, the advanced teers were recruited to undergo MRI scans of the optic non-PTX might be more applicable in clinical practice nerve. All volunteers received and signed informed con- because there is no need for parallel transmission con- sent before undergoing routine MRI and DWI examina- figuration [22]. The majority of imaging parameters of tion of the orbit. All subjects were aged 18 years or older rFOV EPI were referred to conventional ss-EPI DWI, and had no history of neurological disorders, amblyo- and the FOV was reduced to 120 × 98  mm , which cov- pia, or optic nerve diseases. Exclusion criteria included ered the whole of the orbits region and the optic chiasma. the following: (1) subjects with contraindication to MRI The inplane resolution was 1.3 × 1.3  mm , resulting in examination (such as pacemaker installation, metal the scan time was almost the same as the ss-EPI DWI. implants in the body, claustrophobia, etc.); (2) the images To achieve inplane resolution was almost the same in rs- cannot be observed due to noticeable motion artifacts, EPI DWI (1.1 × 1.1  mm ) and rFOV EPI, readout partial susceptibility artifacts, etc. Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 3 of 11 Fourier of 5/8 and readout segments of 7 were applied in were defined to quantify the geometric distortion. The rs-EPI DWI, but with the longest scan time of 3:12 min. length and the maximum angle of the intraorbital seg- The specific parameters of three DWI sequences are ment of the optic nerve were measured on each side of listed in Table 1. the eye on this slice which covered the largest extent of the optic nerve of the EPI sequences with a b-value Evaluation and analysis of 1000  s/mm ; T2w imaging was used as a reference. ADC values The quantitative analysis was performed using RadiAnt The ADC values of bilateral optic nerves were assessed Dicom Viewer (version 5.5.0). An example is shown in on the ADC maps at slice that covered the largest extent Fig. 1. of the optic nerves from ss-EPI DWI, rFOV DWI, and rs- EPI DWI images. The measurements were made in each Image quality of the two repeated scans to evaluate the reproducibility. We compared the image quality of rs-EPI DWI, rFOV Three regions of interest (ROIs) using the shape of a cir - DWI, and ss-EPI DWI using a 5-point scale visual evalu- cle or ellipse and encompassing an area of 50 (± 3) mm ation. Two radiological technologists with 4 and 7  years were measured on each side, as shown in Fig.  1. ROI s of experience independently evaluated image qual- were performed on the starting point of the intraorbi- ity in the qualitative assessments of the four aspects on tal segment of the optic nerve and were centered within the b = 1000  s/mm images; The FOV of the DWIs was 3  mm of the back of the eyeball. ROI s were performed adapted so that the readers were blinded to the used on the end of the intraorbital segment of the optic nerve. sequence. The detailed evaluation criteria of image qual - The center of ROI s was at the midpoint of the line ity are shown in Table  2. Mean scores were calculated between the center of ROI s and the center of ROI s. ROI from those of the two readers. Similarly, the signal-to- 1 3 was drawn on the image of the first scan and sent to that noise ratio of images was measured. However, due to of the second scan for the same sequence. Each ROI was the distortion of parallel acquisition sequence, as well measured three times, and the means were recorded as a as the lack of inclusion of background air in the rFOV result. DWI images, consistent with a previous report regarding estimating SNR in the presence of parallel imaging, the Distortion artifacts analysis estimated signal-to-noise ratio (eSNR, the ratio between The length of the intraorbital segment of the optic nerve the mean and standard deviation (SD) of the given ROI.) and the maximum angle at which the optic nerve bends corresponding to each ROI on the b-1000 images was Table 1 Magnetic resonance imaging sequence parameters T2w imaging rs-EPI DWI rFOV DWI ss-EPI DWI TR (ms) 4400 4650 3500 3700 TE (ms) 99 68 65 80 FOV (mm ) 180 × 180 180 × 108 120 × 98 210 × 210 Voxel size (mm ) 0.5 × 0.5 × 2.5 1.1 × 1.1 × 2.5 1.3 × 1.3 × 2.5 1.8 × 1.8 × 2.5 b-value (s/mm ) – 0, 1000 0, 1000 0, 1000 Diffusion directions – 3 3 3 Averages per b-value – 2,2 2,10 2,10 PAT mode GRAPPA GRAPPA GRAPPA GRAPPA Acceleration factor PE 2 2 2 2 In-plane resolution (mm ) 0.5 × 0.5 1.1 × 1.1 1.3 × 1.3 1.8 × 1.8 Slice thickness (mm) 2.5 2.5 2.5 2.5 Base resolution 384 160 96 120 Echo spacing (mm) 11.0 0.36 0.99 0.76 Bandwidth (Hz/px) 250 679 1158 1488 EPI factor – 96 78 120 B0 Shim mode Standard Standard Standard Standard TA (min) 1:47 3:12 2:03 2:11 rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging; TE, echo time; TR, repetition time, TA, acquisition time, GRAPPA: generalized autocalibrating partially parallel acquisitions Zhou et al. BMC Medical Imaging (2022) 22:96 Page 4 of 11 Fig. 1 Examples of quantitative optic nerve measurements. a Measurement of ADC values. Three ROIs were measured on each side. b, Measurement of ADC values. Three ROIs were measured on the left side, along with (left to right) images of ADC map of rs-EPI DWI, ADC map of rFOV-EPI DWI and ADC map of ss-EPI DWI, respectively; c the length of the intraorbital segment of the optic nerve is shown, along with (left to right) 2 2 2 images of T2w, rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm (respectively); d Maximum angles of the optic nerve are shown, along with (left to right) images of T2w, rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b 2 2 value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm (respectively). Compared with the T2w images, the DWI images had different degrees of distortion, and the angle of the optic nerve became smaller. ADC, apparent diffusion coefficient; rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging Table 2 Evaluation criteria of image quality Overall image quality Artifacts (degree of Image blurring Distortion (compared with susceptibility) the corresponding T2w imaging) 0 Non-diagnostic No artifacts None None 1 Poor Minor artifacts Minimal Minimal 2 Fair Moderate artifacts Moderate Moderate 3 Good Severe artifacts Considerable Considerable 4 Excellent Very severe artifacts Pronounced Pronounced calculated [23]. The ROIs were placed on an area show - (Chicago, IL, USA). A P value of less than 0.05 was con- ing visually normal signal intensity and an absence of sidered statistically significant. artifacts (The ROIs of the three sequences were placed in Shapiro–Wilk was used to check the normal distribu- the grey matter region of the left frontal lobe at the same tion of the data. The reproducibility of the mean ADC slice) and had a mean size of 50 (± 3) mm . The mean and of two repeated MR examinations was established using SD of all ROIs were recorded. 95% Bland–Altman limits of agreements (BA-LA) and intraclass correlation coefficients (ICCs, two-way ran - Statistical analysis dom). ANOVA or Friedman test (depending on whether The statistical data were analyzed using MedCalc 11.5.0 fulfilling the assumption of the parametric test) was (https:// www. medca lc. org/) and SPSS version 22.0 performed to compare the mean ADC values and eSNR Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 5 of 11 of three sequences. In addition, a two-tailed pair Stu- ROI and ROI . For rs-EPI DWI, the mean ADC value 2 3 dent’s t-test was used to compare the difference of mean of the middle segment of the optic nerve was relatively ADC values between each pair of left and right nerves, unstable, and the reproducibility was the lowest. In con- respectively. trast, the reproducibility of the mean ADC value of ROI The agreement of the length of the intraorbital seg - was significantly higher than that of ROI and ROI for 2 3 ment of the optic nerve and the maximum angle at which all three DWI sequences. In addition, the reproducibility the optic nerve bends of the three techniques with T2w of the mean ADC value of ROI and ROI3 of rs-EPI and imaging was determined by the calculation of ICCs (two- ROI of rFOV-DWI were good (ICCs > 0.75). way random). In the qualitative assessments of the five aspects of image quality, the mean scores of the two read- ADC values ers were calculated. Results are presented as mean ± SD. There were no statistically significant differences in mean Bonferroni correction for multiple comparisons was per- ADC values between the ROIs in the left and right optic formed after the overall significance was achieved. All nerve from the three techniques (P > 0.05). Therefore, aspects of image quality were evaluated using the Wil- the ADC values from the three methods were compared coxon signed-rank test. using the average ADC values of the left and right optic nerves of each ROI. Results For each DWI technique, mean ADC values among Twenty-six volunteers (6 men, 20 women; mean age, the three different ROIs showed statistically signifi - 31.2  years; range, 22–55  years) of the total 33 were per- cant differences (P < 0.01). In all three DWI sequences, formed the following parameter and image quality evalu- the mean ADC value of ROI was the highest, while the ations (seven volunteers were excluded, four volunteers mean ADC value of ROI was the lowest (Table  4). Mul- dropped out of the experiment due to personal reasons tiple comparisons showed the following: a) there were and three were due to severe motion artifacts and sus- significant differences between ROI and ROI /ROI in 1 2 3 ceptibility artifacts of images). all three DWI methods (P < 0.01), and b) there were no significant differences between ROI and ROI in rs-EPI 2 3 Reproducibility DWI (P = 0.294), rFOV DWI (P = 0.770) and ss-EPI DWI The results from the 95% Bland–Altman limits of agree - (P = 0.162). ments (BA-LA) and intraclass correlation coefficients For each ROI, mean ADC values among the three dif- (ICCs, two-way random) are shown in Table  3. Bland– ferent DWI techniques showed statistically significant Altman plots are also given to show reproducibility of differences (P < 0.001). In all three ROIs, the mean ADC mean ADC values in the optic nerve during repeat MRI value of rs-EPI DWI was the highest, while the mean exams (Fig. 2). rs-EPI DWI showed the highest reproduc- ADC value of rFOV DWI was the lowest (Table 4). Mean ibility of mean ADC value in three ROIs (ICCs > 0.75), ADC values of ROI (P = 0.058), ROI (P = 0.315) and 1 2 especially ROI (ICCs = 0.916). In comparison, ss-EPI ROI (P = 0.679) between rs-EPI DWI and ss-EPI DWI DWI showed the lowest reproducibility, especially in did not differ significantly. The mean ADC values of rFOV DWI were significantly lower than those of rs-EPI DWI and ss-EPI DWI in all three ROIs. Table 3 Reproducibility of ADC Values in repeated MR exams Image quality rs-EPI rFOV-DWI ss-EPI Using data pooled between the two technologists, the ROI overall image quality of the DW images of rFOV DWI ICCs 0.916 0.850 0.810 and rs-EPI DWI were significantly higher (P < 0.001) than BA-LA − 20 to 22 − 18 to 28.1 − 33.5 to 22.2 the image quality of the ss-EPI DWI images (Fig. 3). Arti- ROI 2 facts were significantly less severe (P < 0.001) in the rFOV ICCs 0.797 0.595 0.442 DWI and rs-EPI DWI sequences compared with ss-EPI BA-LA − 31 to 32.2 − 26.5 to 29 − 46.2 to 51.1 DWI. Likewise, distortion (P < 0.001) and image blur- ROI 3 ring (P < 0.001) of the DW images were rated significantly ICCs 0.781 0.750 0.379 lower for the rFOV DWI and rs-EPI DWI sequence than BA-LA − 40.6 to 18.1 − 19.6 to 25.3 − 41.5 to 41.3 ss-EPI DWI (Fig.  3). The results are shown in Table  5. Pairwise comparisons revealed significant differences BA-LA (in %), the 95% Bland–Altman limits of agreements; ICCs, intraclass correlation coefficients (two-way random); rs-EPI DWI, readout-segmented echo- between ss-EPI DWI and the other two sequences in planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, image quality, artifacts, distortion, and image blur- single-shot echo-planar imaging; ROI, region of interest ring, with P values of P = 0.001, P = 0.039, P < 0.001, and P > 0.05 Zhou et al. BMC Medical Imaging (2022) 22:96 Page 6 of 11 Fig. 2 Reproducibility and accuracy of ADC values assessment. Bland–Altman plots show reproducibility of mean ADC values in the optic nerve during repeat exams. rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging. a1–a3 Reproducibility evaluation of mean ADC values of rs-EPI DWI. a1–a3 are the Bland–Altman plots of R OI , ROI , and ROI , respectively. b1–b3 Reproducibility evaluation of mean ADC values of rFOV DWI. b1–b3 are the Bland–Altman plots of R OI, ROI , 2 3 1 2 and ROI , respectively. c1–c3 Reproducibility evaluation of mean ADC values of ss-EPI DWI. c1–c3 are the Bland–Altman plots of R OI, ROI , and 3 1 2 ROI , respectively. In each plot, the solid line represents the average value (percentage) of the difference between the two scans; the dashed lines represent the 95% CI (confidence intervals) of the difference; the blue error bars close to the y-axis represent 95% CI of upper limit, 95% CI of arithmetic mean, 95% CI of the lower limit, respectively P < 0.001 (respectively) versus rs-EPI DWI; and P < 0.001, Distortion artifacts analysis P = 0.012, P = 0.009, P < 0.001, and P < 0.001 versus rFOV Results of distortion artifacts analysis are shown in DWI. For the comparison of rs-EPI DWI and rFOV Table  6. Regarding the anatomic agreement of the three DWI, there were significant differences in image blur - EPI techniques assessed by ICCs, rs-EPI DWI and rFOV ring (P = 0.002) and distortion (P < 0.001), but the differ - DWI showed stronger agreement (higher ICCs) than did ences were not statistically significant for other image ss-EPI DWI for both the length of the optic nerve and the quality parameters. In terms of eSNR, there were statis- maximum angle of the optic nerve. And the agreement of tically significant differences in eSNR among three DWI the maximum angle of the optic nerve was stronger than sequences. eSNR of rFOV DWI images (17.78 ± 4.16) was that of the length of the optic nerve in all DWI methods. lower than that of ss-EPI DWI (26.77 ± 8.58), but was The differences of the quantitative measurements in rela - higher than that of rs-EPI DWI (10.18 ± 1.89), each with tion to T2w imaging (mean(range)) varied in the length P < 0.01. of the left optic nerve (rs-EPI: 0.26 (0.01–0.87)  mm, Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 7 of 11 −6 2 Table 4 Results of quantitative evaluations of mean ADC values (10 × mm /s) ROI ROI ROI F/χ P 1 2 3 First d d c e rs-EPI DWI 1939.79 ± 432.73 1344.62 ± 278.79 1043.69 ± 199.62 43.00 < 0.001 d d d f rFOV-DWI 1251.02 ± 209.81 998.69 ± 168.511 995.99 ± 139.73 18.20 < 0.001 d d d e ss-EPI DWI 1744.47 ± 413.35 1271.74 ± 310.28 1088.33 ± 169.14 36.54 < 0.001 2 e e e F/χ 27.77 21.77 9.80 P < 0.001 < 0.001 0.008 Second d d d f rs-EPI DWI 1908.60 ± 337.60 1335.59 ± 284.62 1249.13 ± 256.82 38.40 < 0.001 c d d e rFOV-DWI 1117.77 ± 277.55 977.81 ± 94.24 965.94 ± 116.86 19.36 < 0.001 d d c e ss-EPI DWI 1675.44 ± 429.45 1225.84 ± 221.47 1034.00 ± 220.31 23.66 < 0.001 2 e e e F/χ 29.96 31.69 24.92 P < 0.001 < 0.001 < 0.001 ADC, apparent diffusion coefficient; rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EP, single-shot echo-planar imaging; ROI, region of interest Mean ADC values of the first scan Mean ADC values of the second scan Parameter value was a non-normal distribution, expressed as a median ± quartile range Parameter value was a normal distribution, expressed as mean ± standard deviation e 2 χ value of Friedman’s test F value of ANOVA rFOV-EPI: 0.27(0–0.98)  mm and ss-EPI :0.31 (0.01– assessment of various diseases, such as trauma [29] and 0.97)  mm), maximum angle of the left optic nerve (rs- ischemia [30, 31], that involve the optic nerves. DWI can EPI: 5.92(0–23.8)°, rFOV-EPI: 6.84 (1.2–20.7)° and ss-EPI: also be considered the first technique capable of identify - 11.7(0.2–58.4)°), length of the right optic nerve (rs-EPI: ing posterior ischemic optic neuropathy (PION) by iden- 0.21(0.03–0.56) mm, rFOV-EPI: 0.25(0.02–0.81)  mm tifying acute ischemic lesions of the optic nerve [32]. And and ss-EPI: 0.34(0.02–0.92) mm), and maximum angle of ADC may serve as a useful tool for prognostication for the right optic nerve (rs-EPI: 7.8(0.7–20.6)°, rFOV-EPI: Optic pathway glioma (OPG), a significantly higher mean 8.88(0.4–25.8)° and ss-EPI: 11.85(0.3–53.5)°). The results ADC was seen in OPG that required therapy for tumor of the difference were consistent with the results of ICCs. progression [33]. The differences between the length of the optic nerve and But DWI of the optic nerve is difficult in clinical cir - the angle and optic nerve of rs-EPI DWI and the T2WI cumstances, on account of the small dimension of the image were the smallest, while geometric distortion was nerves, uncontrolled eye movements, and the high sig- more obvious on the ss-EPI DWI images. Compared with nal from cerebrospinal fluid or neighbouring fat within the rs-EPI and rFOV-EPI, the length of the right optic the orbital region [8]. ss-EPI DWI suffers from magnetic nerve of ss-EPI DWI images was significantly shorter, susceptibility artifacts, chemical shift artifacts, and low and the angle of inward bending the middle and posterior image quality. The present study found that rFOV DWI optic nerve was larger (Fig. 3). exhibited superior performance compared with ss-EPI DWI in all evaluated aspects, including blurring effects, Discussion image distortion, artifacts, lesion conspicuity, and image Since DWI has been proven to be a critical functional quality. Barker et al. also found that rFOV DWI provided technology in detecting and locating optic nerve dis- improved subjective image quality of optic neuritis com- ease lesions, for example, the use of DWI and the calcu- pared with ss-EPI DWI [34]. Owing to the readout-seg- lation of ADC values for evaluating optic neuritis (ON) mented k-space acquisition strategy, rs-EPI effectively has been reported [24, 25]. Using only DWI has high reduces the image distortion caused by the large suscep- sensitivity and specificity in distinguishing acute from tibility variations and the T2* blur effect [35]. Therefore, chronic ON. Differences in ADC values can reflect dif - the application of ss-EPI in optic nerve imaging has been ferent pathogenesis of ON [26], and ADC values give a limited in the past. In this study, we evaluated new tech- measure of axonal disruption in the chronic optic nerve nical approaches of DWI of the optic nerve. lesion [27, 28]. Moreover, DWI has been used for the Zhou et al. BMC Medical Imaging (2022) 22:96 Page 8 of 11 Fig. 3 Examples of image quality and distortion of three DW techniques. a Imaging in a 28-year-old female volunteer. a1–a4 images of T2w, 2 2 2 rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm , respectively. b Imaging in a 2 2 25-year-old female volunteer. b1–b4 images of T2w, rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm , respectively. Note the improved image quality and markedly reduced susceptibility distortions in the rs-EPI and rFOV-EPI sequences compared with ss-EPI. There are severe distortions due to the susceptibility artifacts caused by the air-bone-tissue interface on ss-EPI DWI (arrows in a4, b4). rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging rs-EPI DWI showed higher reproducibility than did ss-EPI DWI and rFOV DWI. In the present study, rs- Table 5 Results of image quality evaluations EPI DWI appeared to be the most robust and reliable method. The rs-EPI arrangement partitions the k-space rs-EPI DWI rFOV DWI ss-EPI DWI P trajectory into numerous portions in the readout direc- Image quality 2.38 ± 0.90 2.46 ± 0.58 1.58 ± 0.64 < 0.001 tion. Accordingly, TE and encoding times can be Artifacts 1.96 ± 0.91 1.85 ± 0.67 2.27 ± 0.67 0.042 decreased, and movement correction can be performed Distortion 1.5 ± 0.58 1.81 ± 0.63 3.04 ± 0.72 < 0.001 utilizing a 2D navigator correcting motion-induced, Image blurring 1.35 ± 0.48 1.73 ± 0.45 3.04 ± 0.53 < 0.001 non-linear phase errors [36, 37]. However, the prob- lem is that the acquisition time is longer on account of rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 9 of 11 Table 6 Results of quantitative evaluations of anatomic usually used as the control region. Still, the ADC value measurements of the area with no obvious abnormal signal is usually measured in time for bilateral involvement as a refer- L-length R-length L-angle (°) R-angle (°) (mm) (mm) ence. ADC measurement of the optic nerve is challeng- ing due to the small diameter of the optic nerve because T2WI 2.44 ± 0.24 2.36 ± 0.18 169.61 ± 7.51 170.58 ± 7.12 there is the potential of partial volume averaging with rs-EPI DWI 2.25 ± 0.23 2.20 ± 0.23 165.24 ± 9.26 166.78 ± 11.02 surrounding CSF, fat, and osseous structures. Suscep- ICCs 0.691 0.734 0.993 0.989 tibility artifacts caused by the air-bone-tissue interface 95%CI 0.341–0.855 0.426–0.877 0.986–0.997 0.977–0.995 also affected the ROI placement over the optic nerve. rFOV DWI 2.21 ± 0.308 2.13 ± 0.28 165.64 ± 8.63 165.82 ± 11.24 Therefore, three ROIs in different positions were placed ICCs 0.687 0.667 0.992 0.985 to explore the possible differences in ADC values. The 95%CI 0.334–0.853 0.281–0.846 0.984–0.996 0.969–0.993 comparison of ADC values shows that the ADC value ss-EPI DWI 2.17 ± 0.32 2.04 ± 0.31 158.92 ± 14.9 161.79 ± 17.72 of ROI has the best reproducibility and is signifi - ICCs 0.659 0.675 0.978 0.971 cantly higher than the ADC values of R OI and R OI , 2 3 95%CI 0.275–0.840 0.297–0.850 0.954–0.990 0.939–0.987 indicating that the signal of the optic nerve behind the rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted global is continuously affected by partial volume aver - imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging; aging with the high signal of the eyeball and needs to L-length, the length of the left optic nerve; R-length, the length of the right optic nerve; L-angle, the maximum angle of the left optic nerve; R-angle, the be taken into consideration to rule out false positives. maximum angle of the right optic nerve And when choosing the control region for comparison, Anatomic agreement of three sequences with T2w imaging, measured by the area close to the eyeball should be avoided. There intraclass correlation coefficients (ICCs) b are both good reproducibility in ADC values of ROI 95% confidence interval (CI) of ICCs. T2WI, T2 weighted imaging and ROI , and the difference between ADC of ROI and 3 2 ROI is not statistically significant, which may indicate multiple TR intervals. Seeger et al. found that rFOV-EPI that compared to ROI, ROI does not suffer more inac - 2 3 (2:45  s) showed improved image quality, the most accu- curate due to susceptibility artifacts caused by the air- rate tumor delineation, and the best differentiation from bone-tissue interface and partial volume averaging with retinal detachments compared with ss-EPI (2:14  s) and the surrounding air and osseous structures. rs-EPI (3:07 s) in patients with uveal melanomas [38]. In Distortion creates a challenge for diffusion MRI of the the present study, the reproducibility of the mean ADC optic nerve, especially at high field strengths. The optic value of rFOV DWI (2:03  s) was better than that of ss- nerve is surrounded by fat, muscle, bone, and air lead- EPI DWI (2:11  s), but the reproducibility of R OI was ing to large susceptibility changes that distort diffusion- lower than for rs-EPI DWI (3:12 s). The mean ADC value weighted EPI images. Our research found that rs-EPI of rFOV DWI was significantly lower than those of rs- DWI showed the highest agreement with T2w imaging in EPI DWI and ss-EPI DWI, which is consistent with the terms of the length and angle of the optic nerve compared results of Seeger et  al. [18]. We speculate the reason for with those obtained with rFOV DWI and ss-EPI. The dis - these findings may be that we applied FOV rotation to tortion in the length of the optic nerve may indicate the remove the potential folding artifacts of rFOV-DWI and degree of compression of the image in the long axis of the used complex averages to improve the ADC estimation. optic nerve. The distortion of the angle of the optic nerve Furth more, rFOV DWI overcomes the major problem may represent the distortion ratio of the x-axis compared of low specific absorption rate and low spatial resolution with the y-axis. Previous studies have reported quantita- facing DWI on the optic nerve as the reproducibility of tive evaluations of the degree of distortion in this pro- the ADC values of the ss-EPI DWI sequence is relatively cess. In the study of Thierfelder et al. [13], the rFOV DWI low. It is not recommended to use the ADC values of the of the prostate showed a stronger correlation with the ss-EPI DWI sequence as an indicator in follow-up cases. T2w images in the coronal and sagittal diameters as well Common diseases of the optic nerve include optic as in the prostate volume, compared with those obtained neuritis, ischemic optic neuropathy, optic nerve by ss-EPI DWI, yielding ICCs of 0.948 for the coronal tumors, etc. The clinical manifestations and involve - diameter, 0.858 for the sagittal diameter, and 0.938 for ment of optic nerve segments in different diseases are the prostate volume. These observations are consistent different. For example, inflammation of the optic nerve (to a certain extent) with the results of the present study. can be involved unilaterally or bilaterally, and it can Specifically, we observed that rFOV DWI has a good cor - also be involved in long segments. In the application relation with T2w images in terms of angle. Still, there of the ADC values of the optic nerve, the ADC value is distortion in the long axis of the optic nerve, which is of the contralateral side of the diseased optic nerve is like the fact that the ICC coefficient of the prostate also is Zhou et al. BMC Medical Imaging (2022) 22:96 Page 10 of 11 GRAPPA: Generalized autocalibrating partially parallel acquisitions; BA-LA (in low in the longitudinal axis direction. We speculate that %): The 95% Bland–Altman limits of agreements. the lower ICC coefficient in the longitudinal axis may be due to the susceptibility artifacts caused by the air-bone- Acknowledgements We thank all volunteers for consenting to take part in this study and Kexu He tissue interface. for lots of advice and suggestions for this study. Author contributions All authors were substantially involved in the conceptual design of the Limitations study. FZ performed the MR examination of the optic nerve, analyzed and There are several limitations in this study. Firstly, our interpreted the patient data, and was a major contributor in writing the manu- quantitative analysis was restricted to a global com- script. QL critically reviewed the manuscript and gave important intellectual input. XZ and HM performed the MR examination of the optic nerve. GZ and parison of the optic nerve length and angle. It might be SD performed the MR examination of the optic nerve and did the magnetic of interest to determine how well other parameters in resonance imaging assessment; LZ substantially contributed to the acquisition rFOV DWI or rs-EPI DWI correlate with T2w imaging. of volunteers and coordinated magnetic resonance imaging. TB gave substan- tial technical input regarding magnetic resonance imaging. ZZ optimized scan Additionally, the present investigation only evaluated parameters for MRI sequences, performed the MR examination of the optic the intraorbital segment of the optic nerve and did not nerve. All authors read and approved the final manuscript. examine other segments of this nerve. We have made the Funding comparison between two product DWI sequences and a We would like to acknowledge the financial support of Chongqing Sci- prototype ZOOMit DWI sequence; there are still some ence and Health Joint Medical Research Projection 2021. Contract grant other techniques to reduce distortion and improve image number: 2021MSXM251. quality that we haven’t compared, like BLADE DWI and Availability of data and materials ss-EPI with two b0 acquisitions of opposite phase-encod- The datasets generated and/or analysed during the current study are not ing directions, etc. Secondly, the methods tested in this publicly available due to the confidentiality statement of informed consent (In informed consent: “Information that could identify you will not be disclosed to study are not yet widely used in clinical practice. So, it members outside the research team unless you have given your permission. seems that they may not be available on the specific ven - All study members and study sponsors are required to keep your identity dor’s platform. Furthermore, the performance results of confidential”). But are available from the corresponding author on reasonable request. the three sequences and the results of the image quality assessment may vary depending on the specific hardware. Declarations Therefore, the findings reported in this study are spe - cific to the hardware of these tests. Thirdly, the number Ethics approval and consent to participate of patients examined was only adequate for an explora- The study was approved by the First Affiliated Hospital of Chongqing Medi- cal University (Ethics Number: 2020-085), Chongqing, China. All procedures tory study; additional work with a larger population and performed in the studies involving human participants were performed in patients with pathology is warranted to corroborate our accordance with the ethical standards laid down in the 1964 Declaration findings. Although ADC values of different methods are of Helsinki. Informed consent was obtained from all individual participants included in the study. compared in volunteers, there is no reference to the gold standard, so ADC phantom can be further used to verify Consent for publication the standard in future experiments. Consent for publication was obtained from all individual participants whose individual person’s data was contained in this manuscript. Competing interests Conclusion The authors declare that they have no competing interests. rs-EPI DWI and rFOV-EPI DWI enable high-quality Author details imaging in the intraorbital segment of the optic nerve. 1 Department of Radiology, The First Affiliated Hospital of Chongqing Medi- Both rs-EPI DWI and rFOV-EPI DWI are suitable tech- cal University, No. 1, Youyi Road, Yuzhong District, Chongqing, China. MR Collaborations, Siemens Healthcare Ltd., Shanghai, China. MR Application niques for the assessment of diffusion restriction and Predevelopment, Siemens Healthcare GmbH, Erlangen, Germany. provide significantly improved image quality compared with ss-EPI DWI. For methods using the same acquisi- Received: 10 January 2022 Accepted: 26 April 2022 tion time, rFOV DWI is superior to ss-EPI DWI, while rs-EPI is the best choice, although this technique took 47% longer to perform. References 1. Abiose A, et al. Reduction in incidence of optic nerve disease with annual ivermectin to control onchocerciasis. Lancet. 1993;341:130–4. Abbreviations 2. Onodera M, Yama N, Hashimoto M, et al. The signal intensity ratio of the ADC: Apparent diffusion coefficient; DWI: Diffusion-weighted imaging; EPI: optic nerve to ipsilateral frontal white matter is of value in the diagnosis Echo-planar imaging; rs-EPI DWI: Readout-segmented echo-planar imaging of acute optic neuritis. Eur Radiol. 2016;26(8):2640–5. diffusion-weighted imaging; rFOV: Reduced field-of-view; ss-EPI: Single-shot echo-planar imaging; FOV: Field of view; ROI: Region of interest; ICC: Intraclass correlation coefficient; TE: Echo time; TR: Repetition time; TA: Acquisition time; Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 11 of 11 3. Fatima Z, Ichikawa T, Ishigame K, et al. Orbital masses: the usefulness of 25. Naismith RT, Xu J, Tutlam NT, et al. Disability in optic neuritis corre- diffusion-weighted imaging in lesion categorization. Clin Neuroradiol. lates with diffusion tensor-derived directional diffusivities. Neurology. 2013;24(2):129–34. 2009;72(7):589–94. 4. Bodanapally UK, Shanmuganathan K, Shin RK, et al. Hyperintense optic 26. Levin MH, Bennett JL, Verkman AS. Optic neuritis in neuromyelitis optica. nerve due to diffusion restriction: diffusion-weighted imaging in trau- Prog Retin Eye Res. 2013;36:159-71. matic optic neuropathy. AJNR Am J Neuroradiol. 2015;36:333–48. 27. Hickman SJ, Wheeler-Kingshott CA, Jones SJ, et al. Optic nerve diffusion 5. Yang M, Zhao J, Song H, et al. Orbital magnetic resonance imaging may measurement from diffusion-weighted imaging in optic neuritis. Am J contribute to the diagnosis of optic nerve lymphoma. Front Neurol. Neuroradiol. 2005;140(2):354–5. 2020;11:301. 28. Fatima Z, Motosugi U, Muhi A, et al. Diffusion-weighted imaging in optic 6. Korn A, Heine C, Bischof F, et al. Diffusion restriction of the optic neuritis. Can Assoc Radiol J. 2012;64(1):51–5. nerve in patients with acute visual deficit. J Magn Reson Imaging. 29. Schwenk A, Kollias SSJEJoRE. Bilateral restricted diffusion in optic nerves 2014;40(2):334–40. as an indicator of traumatic optic neuropathy. 2007;64(3):87–89. 7. Fatima Z, Motosugi U, et al. Diffusion-weighted imaging in optic neuritis. 30. Al-Shafai LS, Mikulis DJ. Diffusion MR imaging in a case of acute ischemic Can Assoc Radiol J. 2013;64:51–5. optic neuropathy. AJNR Am J Neuroradiol.2006;27(2):255–7. 8. Purohit BS, Vargas MI, Ailianou A, et al. Orbital tumours and tumour-like 31. Al-Zubidi N, Stevens S, Fung SH, et al. Diffusion-weighted imaging in pos- lesions: exploring the armamentarium of multiparametric imaging. terior ischemic optic neuropathy. Can J Ophthalmol. 2014;49(1):e21–e25. Insights Imaging. 2016;7:43–68. 32. Bender B, Heine C, et al. Diffusion restriction of the optic nerve in patients 9. Sapkota N, Shi X, et al. Two-dimensional single-shot diffusion-weighted with acute visual deficit. J Magn Reson Imaging. 2013;40:334–40. stimulated EPI with reduced FOV for ultrahigh-b radial diffusion-weighted 33. Yeom KW, Lober RM, Andre JB, et al. Prognostic role for diffusion- imaging of spinal cord. Magn Reson Med. 2017;77:2167–73. weighted imaging of pediatric optic pathway glioma. J Neurooncol. 10. Seeger A, Klose U, Bischof F, et al. Zoomed EPI DWI of acute spinal 2013;113(3):479–83. ischemia using a parallel transmission system. Clin Neuroradiol. 34. Barker GJ. Diffusion-weighted imaging of the spinal cord and optic nerve. 2016;26(2):177–82. J Neurol Sci. 2001;186(supp-S1):S45–9. 11. Mannelli L, Monti S, Corrias G, et al. Comparison of navigator triggering 35. Xu XQ, Liu J, Hu H, et al. Improve the image quality of orbital 3 T diffusion- reduced field of view and large field of view diffusion-weighted imaging weighted magnetic resonance imaging with readout-segmented echo- of the pancreas. J Comput Assist Tomogr. 2018;43(1):1. planar imaging. Clin Imaging. 2016;40(4):793–6. 12. Brendle C, Martirosian P, Schwenzer NF, et al. Diffusion-weighted imaging 36. Porter DA, et al. High resolution diffusion-weighted imaging using read- in the assessment of prostate cancer: comparison of zoomed imaging out-segmented echo-planar imaging, parallel imaging and a two-dimen- and conventional technique. Eur J Radiol. 2016;85(5):893–900. sional navigator-based reacquisition. Magn Reson Med. 2009;62:468–75. 13. Thierfelder KM, Scherr MK, Notohamiprodjo M, et al. Diffusion-weighted 37. Cohen-Adad J. High-resolution DWI in brain and spinal cord with syngo MRI of the prostate: advantages of Zoomed EPI with parallel- RESOLVE1. Siemens Magnetom-Clin Neurol. 2012;510:16–23. transmit-accelerated 2D-selective excitation imaging. Eur Radiol. 38. Seeger A, Batra M, Süsskind D, et al. Assessment of uveal melanomas 2014;24(12):3233–41. using advanced diffusion-weighted imaging techniques: value of 14. Dong H, Li Y, Li H, et al. Study of the reduced field-of-view diffusion- reduced field of view DWI (“zoomed DWI”) and readout-segmented DWI weighted imaging of the breast. Clin Breast Cancer. 2014;14(4):265–71. (RESOLVE). Acta Radiol. 2018;60:977–84. 15. Song M-H, Jin Y-F, et al. Application of whole-lesion intravoxel incoherent motion analysis using iZOOM DWI to differentiate malignant from benign Publisher’s Note thyroid nodules. Acta Radiol. 2018;60:1127–34. Springer Nature remains neutral with regard to jurisdictional claims in pub- 16. Yuan T, Wang J, et al. Comparison of field-of-view optimized and lished maps and institutional affiliations. constrained undistorted single-shot diffusion-weighted imaging and conventional diffusion-weighted imaging of optic nerve and chiasma at 3T. Neuroradiology. 2018;60(9):903–12. 17. Feng Z, Min X, Sah VK, et al. Comparison of field-of-view (FOV ) optimized and constrained undistorted single shot (FOCUS) with conventional DWI for the evaluation of prostate cancer. Clin Imaging. 2015;39:851–5. 18. Seeger A, Schulze M, Schuettauf F, et al. Advanced diffusion-weighted imaging in patients with optic neuritis deficit—value of reduced field of view DWI and readout-segmented DWI. Neuroradiol J. 2018;31:126–32. 19. Sha Y, Tian G, et al. Diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging, and two-dimensional navigator- based reacquisition in detecting acute optic neuritis. J Magn Reson Imaging. 2016;43:655–60. 20. Chang P, Nassirpour S, Henning AJ. Modeling real shim fields for very high degree (and order) B0 shimming of the human brain at 9.4 T. Magn Reson Med. 2018;79:529–40. 21. Porter DA, Heidemann RM. High resolution diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging and Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : a two-dimensional navigator-based reacquisition. Magn Reson Med. 2010;62(2):468–75. fast, convenient online submission 22. Hu L, Wei L, Wang S, et al. Better lesion conspicuity translates into thorough peer review by experienced researchers in your field improved prostate cancer detection: comparison of non-parallel- transmission-zoomed-DWI with conventional-DWI. Abdom Radiol (N Y ). rapid publication on acceptance 2021;46(12):5659–68. https:// doi. org/ 10. 1007/ s00261- 021- 03268-5. support for research data, including large and complex data types 23. Rosenkrantz AB, Chandarana H, Pfeuffer J, et al. Zoomed echo-planar • gold Open Access which fosters wider collaboration and increased citations imaging using parallel transmission: impact on image quality of diffusion-weighted imaging of the prostate at 3T. Abdom Imaging. maximum visibility for your research: over 100M website views per year 2015;40(1):120–6. 24. Kolbe S, Chapman C, Nguyen T, et al. Optic nerve diffusion changes and At BMC, research is always in progress. atrophy jointly predict visual dysfunction after optic neuritis. Neuroim- Learn more biomedcentral.com/submissions age. 2009;45(3):679–86. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Medical Imaging Springer Journals

Reproducibility and feasibility of optic nerve diffusion MRI techniques: single-shot echo-planar imaging (EPI), readout-segmented EPI, and reduced field-of-view diffusion-weighted imaging

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/reproducibility-and-feasibility-of-optic-nerve-diffusion-mri-0TE00eziBE

References (43)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
1471-2342
DOI
10.1186/s12880-022-00814-5
Publisher site
See Article on Publisher Site

Abstract

Background: Diffusion-weighted imaging (DWI) is an essential technique for optic nerve diseases. However, the image quality of optic nerve DWI is decreased by the distortions and artifacts associated with conventional tech- niques. In order to establish this method as a critical tool in optic nerve diseases, reproducibility and feasibility of new technical and conventional approaches of DWI need to be systematically investigated. Methods: DWIs were acquired using ss-EPI, readout-segmented EPI (rs-EPI) DWI, and reduced field-of-view (rFOV ) DWI. 26 volunteers (mean age 31.2 years) underwent repeated MRI examinations in order to assess scan–rescan reproducibility and accuracy. The apparent diffusion coefficient (ADC) values (three ROIs were measured on each side) were determined to evaluate the reproducibility of each sequence and the differences between the three techniques. To quantify the geometric distortion artifacts, the length of optic nerve and the maximum angle of optic nerve were defined and compared to T2-weighted imaging. In addition, two readers evaluated four different aspects of image quality on 5-point Likert scales. Results: rs-EPI DWI (ICCs: 0.916, 0.797 and 0.781) and rFOV DWI (ICCs: 0.850, 0.595 and 0.750) showed higher repro- ducibility (ICCs: ROI, ROI and ROI ) of mean ADC value in all three ROIs than ss-EPI DWI (ICCs: 0.810, 0.442 and 0.379). 1 2 3 The quantitative analysis of geometric distortion yielded a higher agreement of both rs-EPI DWI and rFOV DWI with T2-weighted imaging than ss-EPI. rs-EPI DWI (2.38 ± 0.90) and rFOV DWI (2.46 ± 0.58) were superior to ss-EPI DWI (1.58 ± 0.64) with respect to overall image quality and other aspects of image quality, each with P < 0.05. The mean ADC values of rFOV DWI were significantly lower than those of rs-EPI DWI and ss-EPI DWI in all three ROIs (P < 0.001). Conclusions: Both rs-EPI DWI and rFOV-EPI DWI are suitable techniques for the assessment of diffusion restriction and provide significantly improved image quality compared with ss-EPI DWI. For methods using the same acquisition *Correspondence: jintianzzw@163.com Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, No. 1, Youyi Road, Yuzhong District, Chongqing, China Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhou et al. BMC Medical Imaging (2022) 22:96 Page 2 of 11 time, rFOV DWI is superior to ss-EPI DWI, while rs-EPI showed an overall superiority, although this technique took 47% longer to perform. Keywords: Optic nerves, Diffusion-weighted imaging, Zoomed EPI, RESOLVE MRI protocol Background MRI was performed with a 3T MR scanner (MAG- Optic nerve diseases cause severe visual disturbances NETOM Skyra, Siemens Healthcare, Erlangen, Ger- that, currently, cannot be diagnostically confirmed on many) using a 20-channel head matrix coil, and gradients ocular examination [1]. Orbital magnetic resonance with a peak amplitude of 45 mT/m and slew rate of imaging (MRI) is a widely utilized method to analyze 200  T/m/s. Imaging sequences included T2w imaging, various optic diseases such as optical neuritis. The wide and DWIs were acquired using ss-EPI, high-resolution application of orbital MRI in clinical also highly val- rs-EPI (readout segmentation of long variable echo ued functional MR technology, particularly diffusion- trains, RESOLVE , Siemens Healthcare), and a proto- weighted imaging (DWI) [2–5]. Single-shot echo-planar type rFOV-EPI (syngo ZOOMit , Siemens Healthcare). imaging (ss-EPI) is a conventional DW technique for The imaging protocol was repeated on each of the sub - rapid assessment of diffusion restriction and was com - jects at an interval of 20–40 min. Volunteers were placed monly used in clinical evaluation and differential diagno - in a comfortable supine position, and a sponge was posi- sis of optic neuritis [6, 7]. But in clinical circumstances, tioned to limit head movement. ss-EPI DWI of optic nerve suffers from partial-volume B0 inhomogeneity can result in a variety of artifacts, effects, magnetic susceptibility artifacts, chemical shift including geometric distortion, image blurring, especially artifacts, and low image quality on account of the small in EPI sequences [20]. For a fair comparison, the standard dimension of the nerves and low resolution [8]. Recent B0 shimming was used in all the sequences; The upside studies have indicated that reduced field-of-view (rFOV) of utilizing modeled shim fields is that there are no issues DWI provides outstanding improvements in image with different FOVs or different resolutions [20]. The quality of the spinal cord [9, 10], pancreas [11], pros- bandwidth was adjusted to the optimal value to make tate [12, 13], breast [14], thyroid nodules [15], and optic the echo spacing close to the minimum value to reduce nerve [16]. Literature has demonstrated that rFOV DWI image distortion in all three DWI sequences. Addition- showed an improvement in subjective image quality for ally, GRAPPA (GeneRalized Autocalibrating Partially optic nerve in the intraorbital segment compared with Parallel Acquisitions) was applied in each of the three ss-EPI [17, 18]. Moreover, Readout-segmented echo- DWI sequences to reduce susceptibility changes at tissue planar imaging (rs-EPI) is a promising technique that has interfaces [21]. already been reported to increase image quality in orbital The rFOV DWI sequence used the ‘Excitation Model’ imaging [19]. Therefore, the present study focused on of ZOOMit, replacing the traditional pulse with slice comparing the reproducibility and feasibility of current selective excitation. Different from conventional zoomed state-of-the-art diffusion techniques for their application technology, the advanced non-parallel transmission in optic nerve analyses. (non-PTX) zoomed-DWI was performed with a rota- tion of the field of excitation (8 deg.) to reduce distortion and echo time and correspondingly improve image qual- Methods ity despite B0 inhomogeneities [22]. In addition, averag- Subjects ing (10 averages) was applied to improve the evaluation From December 2019 to April 2020, 33 healthy volun- accuracy of the ADC values. Moreover, the advanced teers were recruited to undergo MRI scans of the optic non-PTX might be more applicable in clinical practice nerve. All volunteers received and signed informed con- because there is no need for parallel transmission con- sent before undergoing routine MRI and DWI examina- figuration [22]. The majority of imaging parameters of tion of the orbit. All subjects were aged 18 years or older rFOV EPI were referred to conventional ss-EPI DWI, and had no history of neurological disorders, amblyo- and the FOV was reduced to 120 × 98  mm , which cov- pia, or optic nerve diseases. Exclusion criteria included ered the whole of the orbits region and the optic chiasma. the following: (1) subjects with contraindication to MRI The inplane resolution was 1.3 × 1.3  mm , resulting in examination (such as pacemaker installation, metal the scan time was almost the same as the ss-EPI DWI. implants in the body, claustrophobia, etc.); (2) the images To achieve inplane resolution was almost the same in rs- cannot be observed due to noticeable motion artifacts, EPI DWI (1.1 × 1.1  mm ) and rFOV EPI, readout partial susceptibility artifacts, etc. Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 3 of 11 Fourier of 5/8 and readout segments of 7 were applied in were defined to quantify the geometric distortion. The rs-EPI DWI, but with the longest scan time of 3:12 min. length and the maximum angle of the intraorbital seg- The specific parameters of three DWI sequences are ment of the optic nerve were measured on each side of listed in Table 1. the eye on this slice which covered the largest extent of the optic nerve of the EPI sequences with a b-value Evaluation and analysis of 1000  s/mm ; T2w imaging was used as a reference. ADC values The quantitative analysis was performed using RadiAnt The ADC values of bilateral optic nerves were assessed Dicom Viewer (version 5.5.0). An example is shown in on the ADC maps at slice that covered the largest extent Fig. 1. of the optic nerves from ss-EPI DWI, rFOV DWI, and rs- EPI DWI images. The measurements were made in each Image quality of the two repeated scans to evaluate the reproducibility. We compared the image quality of rs-EPI DWI, rFOV Three regions of interest (ROIs) using the shape of a cir - DWI, and ss-EPI DWI using a 5-point scale visual evalu- cle or ellipse and encompassing an area of 50 (± 3) mm ation. Two radiological technologists with 4 and 7  years were measured on each side, as shown in Fig.  1. ROI s of experience independently evaluated image qual- were performed on the starting point of the intraorbi- ity in the qualitative assessments of the four aspects on tal segment of the optic nerve and were centered within the b = 1000  s/mm images; The FOV of the DWIs was 3  mm of the back of the eyeball. ROI s were performed adapted so that the readers were blinded to the used on the end of the intraorbital segment of the optic nerve. sequence. The detailed evaluation criteria of image qual - The center of ROI s was at the midpoint of the line ity are shown in Table  2. Mean scores were calculated between the center of ROI s and the center of ROI s. ROI from those of the two readers. Similarly, the signal-to- 1 3 was drawn on the image of the first scan and sent to that noise ratio of images was measured. However, due to of the second scan for the same sequence. Each ROI was the distortion of parallel acquisition sequence, as well measured three times, and the means were recorded as a as the lack of inclusion of background air in the rFOV result. DWI images, consistent with a previous report regarding estimating SNR in the presence of parallel imaging, the Distortion artifacts analysis estimated signal-to-noise ratio (eSNR, the ratio between The length of the intraorbital segment of the optic nerve the mean and standard deviation (SD) of the given ROI.) and the maximum angle at which the optic nerve bends corresponding to each ROI on the b-1000 images was Table 1 Magnetic resonance imaging sequence parameters T2w imaging rs-EPI DWI rFOV DWI ss-EPI DWI TR (ms) 4400 4650 3500 3700 TE (ms) 99 68 65 80 FOV (mm ) 180 × 180 180 × 108 120 × 98 210 × 210 Voxel size (mm ) 0.5 × 0.5 × 2.5 1.1 × 1.1 × 2.5 1.3 × 1.3 × 2.5 1.8 × 1.8 × 2.5 b-value (s/mm ) – 0, 1000 0, 1000 0, 1000 Diffusion directions – 3 3 3 Averages per b-value – 2,2 2,10 2,10 PAT mode GRAPPA GRAPPA GRAPPA GRAPPA Acceleration factor PE 2 2 2 2 In-plane resolution (mm ) 0.5 × 0.5 1.1 × 1.1 1.3 × 1.3 1.8 × 1.8 Slice thickness (mm) 2.5 2.5 2.5 2.5 Base resolution 384 160 96 120 Echo spacing (mm) 11.0 0.36 0.99 0.76 Bandwidth (Hz/px) 250 679 1158 1488 EPI factor – 96 78 120 B0 Shim mode Standard Standard Standard Standard TA (min) 1:47 3:12 2:03 2:11 rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging; TE, echo time; TR, repetition time, TA, acquisition time, GRAPPA: generalized autocalibrating partially parallel acquisitions Zhou et al. BMC Medical Imaging (2022) 22:96 Page 4 of 11 Fig. 1 Examples of quantitative optic nerve measurements. a Measurement of ADC values. Three ROIs were measured on each side. b, Measurement of ADC values. Three ROIs were measured on the left side, along with (left to right) images of ADC map of rs-EPI DWI, ADC map of rFOV-EPI DWI and ADC map of ss-EPI DWI, respectively; c the length of the intraorbital segment of the optic nerve is shown, along with (left to right) 2 2 2 images of T2w, rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm (respectively); d Maximum angles of the optic nerve are shown, along with (left to right) images of T2w, rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b 2 2 value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm (respectively). Compared with the T2w images, the DWI images had different degrees of distortion, and the angle of the optic nerve became smaller. ADC, apparent diffusion coefficient; rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging Table 2 Evaluation criteria of image quality Overall image quality Artifacts (degree of Image blurring Distortion (compared with susceptibility) the corresponding T2w imaging) 0 Non-diagnostic No artifacts None None 1 Poor Minor artifacts Minimal Minimal 2 Fair Moderate artifacts Moderate Moderate 3 Good Severe artifacts Considerable Considerable 4 Excellent Very severe artifacts Pronounced Pronounced calculated [23]. The ROIs were placed on an area show - (Chicago, IL, USA). A P value of less than 0.05 was con- ing visually normal signal intensity and an absence of sidered statistically significant. artifacts (The ROIs of the three sequences were placed in Shapiro–Wilk was used to check the normal distribu- the grey matter region of the left frontal lobe at the same tion of the data. The reproducibility of the mean ADC slice) and had a mean size of 50 (± 3) mm . The mean and of two repeated MR examinations was established using SD of all ROIs were recorded. 95% Bland–Altman limits of agreements (BA-LA) and intraclass correlation coefficients (ICCs, two-way ran - Statistical analysis dom). ANOVA or Friedman test (depending on whether The statistical data were analyzed using MedCalc 11.5.0 fulfilling the assumption of the parametric test) was (https:// www. medca lc. org/) and SPSS version 22.0 performed to compare the mean ADC values and eSNR Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 5 of 11 of three sequences. In addition, a two-tailed pair Stu- ROI and ROI . For rs-EPI DWI, the mean ADC value 2 3 dent’s t-test was used to compare the difference of mean of the middle segment of the optic nerve was relatively ADC values between each pair of left and right nerves, unstable, and the reproducibility was the lowest. In con- respectively. trast, the reproducibility of the mean ADC value of ROI The agreement of the length of the intraorbital seg - was significantly higher than that of ROI and ROI for 2 3 ment of the optic nerve and the maximum angle at which all three DWI sequences. In addition, the reproducibility the optic nerve bends of the three techniques with T2w of the mean ADC value of ROI and ROI3 of rs-EPI and imaging was determined by the calculation of ICCs (two- ROI of rFOV-DWI were good (ICCs > 0.75). way random). In the qualitative assessments of the five aspects of image quality, the mean scores of the two read- ADC values ers were calculated. Results are presented as mean ± SD. There were no statistically significant differences in mean Bonferroni correction for multiple comparisons was per- ADC values between the ROIs in the left and right optic formed after the overall significance was achieved. All nerve from the three techniques (P > 0.05). Therefore, aspects of image quality were evaluated using the Wil- the ADC values from the three methods were compared coxon signed-rank test. using the average ADC values of the left and right optic nerves of each ROI. Results For each DWI technique, mean ADC values among Twenty-six volunteers (6 men, 20 women; mean age, the three different ROIs showed statistically signifi - 31.2  years; range, 22–55  years) of the total 33 were per- cant differences (P < 0.01). In all three DWI sequences, formed the following parameter and image quality evalu- the mean ADC value of ROI was the highest, while the ations (seven volunteers were excluded, four volunteers mean ADC value of ROI was the lowest (Table  4). Mul- dropped out of the experiment due to personal reasons tiple comparisons showed the following: a) there were and three were due to severe motion artifacts and sus- significant differences between ROI and ROI /ROI in 1 2 3 ceptibility artifacts of images). all three DWI methods (P < 0.01), and b) there were no significant differences between ROI and ROI in rs-EPI 2 3 Reproducibility DWI (P = 0.294), rFOV DWI (P = 0.770) and ss-EPI DWI The results from the 95% Bland–Altman limits of agree - (P = 0.162). ments (BA-LA) and intraclass correlation coefficients For each ROI, mean ADC values among the three dif- (ICCs, two-way random) are shown in Table  3. Bland– ferent DWI techniques showed statistically significant Altman plots are also given to show reproducibility of differences (P < 0.001). In all three ROIs, the mean ADC mean ADC values in the optic nerve during repeat MRI value of rs-EPI DWI was the highest, while the mean exams (Fig. 2). rs-EPI DWI showed the highest reproduc- ADC value of rFOV DWI was the lowest (Table 4). Mean ibility of mean ADC value in three ROIs (ICCs > 0.75), ADC values of ROI (P = 0.058), ROI (P = 0.315) and 1 2 especially ROI (ICCs = 0.916). In comparison, ss-EPI ROI (P = 0.679) between rs-EPI DWI and ss-EPI DWI DWI showed the lowest reproducibility, especially in did not differ significantly. The mean ADC values of rFOV DWI were significantly lower than those of rs-EPI DWI and ss-EPI DWI in all three ROIs. Table 3 Reproducibility of ADC Values in repeated MR exams Image quality rs-EPI rFOV-DWI ss-EPI Using data pooled between the two technologists, the ROI overall image quality of the DW images of rFOV DWI ICCs 0.916 0.850 0.810 and rs-EPI DWI were significantly higher (P < 0.001) than BA-LA − 20 to 22 − 18 to 28.1 − 33.5 to 22.2 the image quality of the ss-EPI DWI images (Fig. 3). Arti- ROI 2 facts were significantly less severe (P < 0.001) in the rFOV ICCs 0.797 0.595 0.442 DWI and rs-EPI DWI sequences compared with ss-EPI BA-LA − 31 to 32.2 − 26.5 to 29 − 46.2 to 51.1 DWI. Likewise, distortion (P < 0.001) and image blur- ROI 3 ring (P < 0.001) of the DW images were rated significantly ICCs 0.781 0.750 0.379 lower for the rFOV DWI and rs-EPI DWI sequence than BA-LA − 40.6 to 18.1 − 19.6 to 25.3 − 41.5 to 41.3 ss-EPI DWI (Fig.  3). The results are shown in Table  5. Pairwise comparisons revealed significant differences BA-LA (in %), the 95% Bland–Altman limits of agreements; ICCs, intraclass correlation coefficients (two-way random); rs-EPI DWI, readout-segmented echo- between ss-EPI DWI and the other two sequences in planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, image quality, artifacts, distortion, and image blur- single-shot echo-planar imaging; ROI, region of interest ring, with P values of P = 0.001, P = 0.039, P < 0.001, and P > 0.05 Zhou et al. BMC Medical Imaging (2022) 22:96 Page 6 of 11 Fig. 2 Reproducibility and accuracy of ADC values assessment. Bland–Altman plots show reproducibility of mean ADC values in the optic nerve during repeat exams. rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging. a1–a3 Reproducibility evaluation of mean ADC values of rs-EPI DWI. a1–a3 are the Bland–Altman plots of R OI , ROI , and ROI , respectively. b1–b3 Reproducibility evaluation of mean ADC values of rFOV DWI. b1–b3 are the Bland–Altman plots of R OI, ROI , 2 3 1 2 and ROI , respectively. c1–c3 Reproducibility evaluation of mean ADC values of ss-EPI DWI. c1–c3 are the Bland–Altman plots of R OI, ROI , and 3 1 2 ROI , respectively. In each plot, the solid line represents the average value (percentage) of the difference between the two scans; the dashed lines represent the 95% CI (confidence intervals) of the difference; the blue error bars close to the y-axis represent 95% CI of upper limit, 95% CI of arithmetic mean, 95% CI of the lower limit, respectively P < 0.001 (respectively) versus rs-EPI DWI; and P < 0.001, Distortion artifacts analysis P = 0.012, P = 0.009, P < 0.001, and P < 0.001 versus rFOV Results of distortion artifacts analysis are shown in DWI. For the comparison of rs-EPI DWI and rFOV Table  6. Regarding the anatomic agreement of the three DWI, there were significant differences in image blur - EPI techniques assessed by ICCs, rs-EPI DWI and rFOV ring (P = 0.002) and distortion (P < 0.001), but the differ - DWI showed stronger agreement (higher ICCs) than did ences were not statistically significant for other image ss-EPI DWI for both the length of the optic nerve and the quality parameters. In terms of eSNR, there were statis- maximum angle of the optic nerve. And the agreement of tically significant differences in eSNR among three DWI the maximum angle of the optic nerve was stronger than sequences. eSNR of rFOV DWI images (17.78 ± 4.16) was that of the length of the optic nerve in all DWI methods. lower than that of ss-EPI DWI (26.77 ± 8.58), but was The differences of the quantitative measurements in rela - higher than that of rs-EPI DWI (10.18 ± 1.89), each with tion to T2w imaging (mean(range)) varied in the length P < 0.01. of the left optic nerve (rs-EPI: 0.26 (0.01–0.87)  mm, Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 7 of 11 −6 2 Table 4 Results of quantitative evaluations of mean ADC values (10 × mm /s) ROI ROI ROI F/χ P 1 2 3 First d d c e rs-EPI DWI 1939.79 ± 432.73 1344.62 ± 278.79 1043.69 ± 199.62 43.00 < 0.001 d d d f rFOV-DWI 1251.02 ± 209.81 998.69 ± 168.511 995.99 ± 139.73 18.20 < 0.001 d d d e ss-EPI DWI 1744.47 ± 413.35 1271.74 ± 310.28 1088.33 ± 169.14 36.54 < 0.001 2 e e e F/χ 27.77 21.77 9.80 P < 0.001 < 0.001 0.008 Second d d d f rs-EPI DWI 1908.60 ± 337.60 1335.59 ± 284.62 1249.13 ± 256.82 38.40 < 0.001 c d d e rFOV-DWI 1117.77 ± 277.55 977.81 ± 94.24 965.94 ± 116.86 19.36 < 0.001 d d c e ss-EPI DWI 1675.44 ± 429.45 1225.84 ± 221.47 1034.00 ± 220.31 23.66 < 0.001 2 e e e F/χ 29.96 31.69 24.92 P < 0.001 < 0.001 < 0.001 ADC, apparent diffusion coefficient; rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EP, single-shot echo-planar imaging; ROI, region of interest Mean ADC values of the first scan Mean ADC values of the second scan Parameter value was a non-normal distribution, expressed as a median ± quartile range Parameter value was a normal distribution, expressed as mean ± standard deviation e 2 χ value of Friedman’s test F value of ANOVA rFOV-EPI: 0.27(0–0.98)  mm and ss-EPI :0.31 (0.01– assessment of various diseases, such as trauma [29] and 0.97)  mm), maximum angle of the left optic nerve (rs- ischemia [30, 31], that involve the optic nerves. DWI can EPI: 5.92(0–23.8)°, rFOV-EPI: 6.84 (1.2–20.7)° and ss-EPI: also be considered the first technique capable of identify - 11.7(0.2–58.4)°), length of the right optic nerve (rs-EPI: ing posterior ischemic optic neuropathy (PION) by iden- 0.21(0.03–0.56) mm, rFOV-EPI: 0.25(0.02–0.81)  mm tifying acute ischemic lesions of the optic nerve [32]. And and ss-EPI: 0.34(0.02–0.92) mm), and maximum angle of ADC may serve as a useful tool for prognostication for the right optic nerve (rs-EPI: 7.8(0.7–20.6)°, rFOV-EPI: Optic pathway glioma (OPG), a significantly higher mean 8.88(0.4–25.8)° and ss-EPI: 11.85(0.3–53.5)°). The results ADC was seen in OPG that required therapy for tumor of the difference were consistent with the results of ICCs. progression [33]. The differences between the length of the optic nerve and But DWI of the optic nerve is difficult in clinical cir - the angle and optic nerve of rs-EPI DWI and the T2WI cumstances, on account of the small dimension of the image were the smallest, while geometric distortion was nerves, uncontrolled eye movements, and the high sig- more obvious on the ss-EPI DWI images. Compared with nal from cerebrospinal fluid or neighbouring fat within the rs-EPI and rFOV-EPI, the length of the right optic the orbital region [8]. ss-EPI DWI suffers from magnetic nerve of ss-EPI DWI images was significantly shorter, susceptibility artifacts, chemical shift artifacts, and low and the angle of inward bending the middle and posterior image quality. The present study found that rFOV DWI optic nerve was larger (Fig. 3). exhibited superior performance compared with ss-EPI DWI in all evaluated aspects, including blurring effects, Discussion image distortion, artifacts, lesion conspicuity, and image Since DWI has been proven to be a critical functional quality. Barker et al. also found that rFOV DWI provided technology in detecting and locating optic nerve dis- improved subjective image quality of optic neuritis com- ease lesions, for example, the use of DWI and the calcu- pared with ss-EPI DWI [34]. Owing to the readout-seg- lation of ADC values for evaluating optic neuritis (ON) mented k-space acquisition strategy, rs-EPI effectively has been reported [24, 25]. Using only DWI has high reduces the image distortion caused by the large suscep- sensitivity and specificity in distinguishing acute from tibility variations and the T2* blur effect [35]. Therefore, chronic ON. Differences in ADC values can reflect dif - the application of ss-EPI in optic nerve imaging has been ferent pathogenesis of ON [26], and ADC values give a limited in the past. In this study, we evaluated new tech- measure of axonal disruption in the chronic optic nerve nical approaches of DWI of the optic nerve. lesion [27, 28]. Moreover, DWI has been used for the Zhou et al. BMC Medical Imaging (2022) 22:96 Page 8 of 11 Fig. 3 Examples of image quality and distortion of three DW techniques. a Imaging in a 28-year-old female volunteer. a1–a4 images of T2w, 2 2 2 rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm , respectively. b Imaging in a 2 2 25-year-old female volunteer. b1–b4 images of T2w, rs-EPI DWI at b value = 1000 s/mm , rFOV-EPI DWI at b value = 1000 s/mm and ss-EPI DWI at b value = 1000 s/mm , respectively. Note the improved image quality and markedly reduced susceptibility distortions in the rs-EPI and rFOV-EPI sequences compared with ss-EPI. There are severe distortions due to the susceptibility artifacts caused by the air-bone-tissue interface on ss-EPI DWI (arrows in a4, b4). rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging rs-EPI DWI showed higher reproducibility than did ss-EPI DWI and rFOV DWI. In the present study, rs- Table 5 Results of image quality evaluations EPI DWI appeared to be the most robust and reliable method. The rs-EPI arrangement partitions the k-space rs-EPI DWI rFOV DWI ss-EPI DWI P trajectory into numerous portions in the readout direc- Image quality 2.38 ± 0.90 2.46 ± 0.58 1.58 ± 0.64 < 0.001 tion. Accordingly, TE and encoding times can be Artifacts 1.96 ± 0.91 1.85 ± 0.67 2.27 ± 0.67 0.042 decreased, and movement correction can be performed Distortion 1.5 ± 0.58 1.81 ± 0.63 3.04 ± 0.72 < 0.001 utilizing a 2D navigator correcting motion-induced, Image blurring 1.35 ± 0.48 1.73 ± 0.45 3.04 ± 0.53 < 0.001 non-linear phase errors [36, 37]. However, the prob- lem is that the acquisition time is longer on account of rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 9 of 11 Table 6 Results of quantitative evaluations of anatomic usually used as the control region. Still, the ADC value measurements of the area with no obvious abnormal signal is usually measured in time for bilateral involvement as a refer- L-length R-length L-angle (°) R-angle (°) (mm) (mm) ence. ADC measurement of the optic nerve is challeng- ing due to the small diameter of the optic nerve because T2WI 2.44 ± 0.24 2.36 ± 0.18 169.61 ± 7.51 170.58 ± 7.12 there is the potential of partial volume averaging with rs-EPI DWI 2.25 ± 0.23 2.20 ± 0.23 165.24 ± 9.26 166.78 ± 11.02 surrounding CSF, fat, and osseous structures. Suscep- ICCs 0.691 0.734 0.993 0.989 tibility artifacts caused by the air-bone-tissue interface 95%CI 0.341–0.855 0.426–0.877 0.986–0.997 0.977–0.995 also affected the ROI placement over the optic nerve. rFOV DWI 2.21 ± 0.308 2.13 ± 0.28 165.64 ± 8.63 165.82 ± 11.24 Therefore, three ROIs in different positions were placed ICCs 0.687 0.667 0.992 0.985 to explore the possible differences in ADC values. The 95%CI 0.334–0.853 0.281–0.846 0.984–0.996 0.969–0.993 comparison of ADC values shows that the ADC value ss-EPI DWI 2.17 ± 0.32 2.04 ± 0.31 158.92 ± 14.9 161.79 ± 17.72 of ROI has the best reproducibility and is signifi - ICCs 0.659 0.675 0.978 0.971 cantly higher than the ADC values of R OI and R OI , 2 3 95%CI 0.275–0.840 0.297–0.850 0.954–0.990 0.939–0.987 indicating that the signal of the optic nerve behind the rs-EPI DWI, readout-segmented echo-planar imaging diffusion-weighted global is continuously affected by partial volume aver - imaging; rFOV, reduced field-of-view; ss-EPI, single-shot echo-planar imaging; aging with the high signal of the eyeball and needs to L-length, the length of the left optic nerve; R-length, the length of the right optic nerve; L-angle, the maximum angle of the left optic nerve; R-angle, the be taken into consideration to rule out false positives. maximum angle of the right optic nerve And when choosing the control region for comparison, Anatomic agreement of three sequences with T2w imaging, measured by the area close to the eyeball should be avoided. There intraclass correlation coefficients (ICCs) b are both good reproducibility in ADC values of ROI 95% confidence interval (CI) of ICCs. T2WI, T2 weighted imaging and ROI , and the difference between ADC of ROI and 3 2 ROI is not statistically significant, which may indicate multiple TR intervals. Seeger et al. found that rFOV-EPI that compared to ROI, ROI does not suffer more inac - 2 3 (2:45  s) showed improved image quality, the most accu- curate due to susceptibility artifacts caused by the air- rate tumor delineation, and the best differentiation from bone-tissue interface and partial volume averaging with retinal detachments compared with ss-EPI (2:14  s) and the surrounding air and osseous structures. rs-EPI (3:07 s) in patients with uveal melanomas [38]. In Distortion creates a challenge for diffusion MRI of the the present study, the reproducibility of the mean ADC optic nerve, especially at high field strengths. The optic value of rFOV DWI (2:03  s) was better than that of ss- nerve is surrounded by fat, muscle, bone, and air lead- EPI DWI (2:11  s), but the reproducibility of R OI was ing to large susceptibility changes that distort diffusion- lower than for rs-EPI DWI (3:12 s). The mean ADC value weighted EPI images. Our research found that rs-EPI of rFOV DWI was significantly lower than those of rs- DWI showed the highest agreement with T2w imaging in EPI DWI and ss-EPI DWI, which is consistent with the terms of the length and angle of the optic nerve compared results of Seeger et  al. [18]. We speculate the reason for with those obtained with rFOV DWI and ss-EPI. The dis - these findings may be that we applied FOV rotation to tortion in the length of the optic nerve may indicate the remove the potential folding artifacts of rFOV-DWI and degree of compression of the image in the long axis of the used complex averages to improve the ADC estimation. optic nerve. The distortion of the angle of the optic nerve Furth more, rFOV DWI overcomes the major problem may represent the distortion ratio of the x-axis compared of low specific absorption rate and low spatial resolution with the y-axis. Previous studies have reported quantita- facing DWI on the optic nerve as the reproducibility of tive evaluations of the degree of distortion in this pro- the ADC values of the ss-EPI DWI sequence is relatively cess. In the study of Thierfelder et al. [13], the rFOV DWI low. It is not recommended to use the ADC values of the of the prostate showed a stronger correlation with the ss-EPI DWI sequence as an indicator in follow-up cases. T2w images in the coronal and sagittal diameters as well Common diseases of the optic nerve include optic as in the prostate volume, compared with those obtained neuritis, ischemic optic neuropathy, optic nerve by ss-EPI DWI, yielding ICCs of 0.948 for the coronal tumors, etc. The clinical manifestations and involve - diameter, 0.858 for the sagittal diameter, and 0.938 for ment of optic nerve segments in different diseases are the prostate volume. These observations are consistent different. For example, inflammation of the optic nerve (to a certain extent) with the results of the present study. can be involved unilaterally or bilaterally, and it can Specifically, we observed that rFOV DWI has a good cor - also be involved in long segments. In the application relation with T2w images in terms of angle. Still, there of the ADC values of the optic nerve, the ADC value is distortion in the long axis of the optic nerve, which is of the contralateral side of the diseased optic nerve is like the fact that the ICC coefficient of the prostate also is Zhou et al. BMC Medical Imaging (2022) 22:96 Page 10 of 11 GRAPPA: Generalized autocalibrating partially parallel acquisitions; BA-LA (in low in the longitudinal axis direction. We speculate that %): The 95% Bland–Altman limits of agreements. the lower ICC coefficient in the longitudinal axis may be due to the susceptibility artifacts caused by the air-bone- Acknowledgements We thank all volunteers for consenting to take part in this study and Kexu He tissue interface. for lots of advice and suggestions for this study. Author contributions All authors were substantially involved in the conceptual design of the Limitations study. FZ performed the MR examination of the optic nerve, analyzed and There are several limitations in this study. Firstly, our interpreted the patient data, and was a major contributor in writing the manu- quantitative analysis was restricted to a global com- script. QL critically reviewed the manuscript and gave important intellectual input. XZ and HM performed the MR examination of the optic nerve. GZ and parison of the optic nerve length and angle. It might be SD performed the MR examination of the optic nerve and did the magnetic of interest to determine how well other parameters in resonance imaging assessment; LZ substantially contributed to the acquisition rFOV DWI or rs-EPI DWI correlate with T2w imaging. of volunteers and coordinated magnetic resonance imaging. TB gave substan- tial technical input regarding magnetic resonance imaging. ZZ optimized scan Additionally, the present investigation only evaluated parameters for MRI sequences, performed the MR examination of the optic the intraorbital segment of the optic nerve and did not nerve. All authors read and approved the final manuscript. examine other segments of this nerve. We have made the Funding comparison between two product DWI sequences and a We would like to acknowledge the financial support of Chongqing Sci- prototype ZOOMit DWI sequence; there are still some ence and Health Joint Medical Research Projection 2021. Contract grant other techniques to reduce distortion and improve image number: 2021MSXM251. quality that we haven’t compared, like BLADE DWI and Availability of data and materials ss-EPI with two b0 acquisitions of opposite phase-encod- The datasets generated and/or analysed during the current study are not ing directions, etc. Secondly, the methods tested in this publicly available due to the confidentiality statement of informed consent (In informed consent: “Information that could identify you will not be disclosed to study are not yet widely used in clinical practice. So, it members outside the research team unless you have given your permission. seems that they may not be available on the specific ven - All study members and study sponsors are required to keep your identity dor’s platform. Furthermore, the performance results of confidential”). But are available from the corresponding author on reasonable request. the three sequences and the results of the image quality assessment may vary depending on the specific hardware. Declarations Therefore, the findings reported in this study are spe - cific to the hardware of these tests. Thirdly, the number Ethics approval and consent to participate of patients examined was only adequate for an explora- The study was approved by the First Affiliated Hospital of Chongqing Medi- cal University (Ethics Number: 2020-085), Chongqing, China. All procedures tory study; additional work with a larger population and performed in the studies involving human participants were performed in patients with pathology is warranted to corroborate our accordance with the ethical standards laid down in the 1964 Declaration findings. Although ADC values of different methods are of Helsinki. Informed consent was obtained from all individual participants included in the study. compared in volunteers, there is no reference to the gold standard, so ADC phantom can be further used to verify Consent for publication the standard in future experiments. Consent for publication was obtained from all individual participants whose individual person’s data was contained in this manuscript. Competing interests Conclusion The authors declare that they have no competing interests. rs-EPI DWI and rFOV-EPI DWI enable high-quality Author details imaging in the intraorbital segment of the optic nerve. 1 Department of Radiology, The First Affiliated Hospital of Chongqing Medi- Both rs-EPI DWI and rFOV-EPI DWI are suitable tech- cal University, No. 1, Youyi Road, Yuzhong District, Chongqing, China. MR Collaborations, Siemens Healthcare Ltd., Shanghai, China. MR Application niques for the assessment of diffusion restriction and Predevelopment, Siemens Healthcare GmbH, Erlangen, Germany. provide significantly improved image quality compared with ss-EPI DWI. For methods using the same acquisi- Received: 10 January 2022 Accepted: 26 April 2022 tion time, rFOV DWI is superior to ss-EPI DWI, while rs-EPI is the best choice, although this technique took 47% longer to perform. References 1. Abiose A, et al. Reduction in incidence of optic nerve disease with annual ivermectin to control onchocerciasis. Lancet. 1993;341:130–4. Abbreviations 2. Onodera M, Yama N, Hashimoto M, et al. The signal intensity ratio of the ADC: Apparent diffusion coefficient; DWI: Diffusion-weighted imaging; EPI: optic nerve to ipsilateral frontal white matter is of value in the diagnosis Echo-planar imaging; rs-EPI DWI: Readout-segmented echo-planar imaging of acute optic neuritis. Eur Radiol. 2016;26(8):2640–5. diffusion-weighted imaging; rFOV: Reduced field-of-view; ss-EPI: Single-shot echo-planar imaging; FOV: Field of view; ROI: Region of interest; ICC: Intraclass correlation coefficient; TE: Echo time; TR: Repetition time; TA: Acquisition time; Zhou  et al. BMC Medical Imaging (2022) 22:96 Page 11 of 11 3. Fatima Z, Ichikawa T, Ishigame K, et al. Orbital masses: the usefulness of 25. Naismith RT, Xu J, Tutlam NT, et al. Disability in optic neuritis corre- diffusion-weighted imaging in lesion categorization. Clin Neuroradiol. lates with diffusion tensor-derived directional diffusivities. Neurology. 2013;24(2):129–34. 2009;72(7):589–94. 4. Bodanapally UK, Shanmuganathan K, Shin RK, et al. Hyperintense optic 26. Levin MH, Bennett JL, Verkman AS. Optic neuritis in neuromyelitis optica. nerve due to diffusion restriction: diffusion-weighted imaging in trau- Prog Retin Eye Res. 2013;36:159-71. matic optic neuropathy. AJNR Am J Neuroradiol. 2015;36:333–48. 27. Hickman SJ, Wheeler-Kingshott CA, Jones SJ, et al. Optic nerve diffusion 5. Yang M, Zhao J, Song H, et al. Orbital magnetic resonance imaging may measurement from diffusion-weighted imaging in optic neuritis. Am J contribute to the diagnosis of optic nerve lymphoma. Front Neurol. Neuroradiol. 2005;140(2):354–5. 2020;11:301. 28. Fatima Z, Motosugi U, Muhi A, et al. Diffusion-weighted imaging in optic 6. Korn A, Heine C, Bischof F, et al. Diffusion restriction of the optic neuritis. Can Assoc Radiol J. 2012;64(1):51–5. nerve in patients with acute visual deficit. J Magn Reson Imaging. 29. Schwenk A, Kollias SSJEJoRE. Bilateral restricted diffusion in optic nerves 2014;40(2):334–40. as an indicator of traumatic optic neuropathy. 2007;64(3):87–89. 7. Fatima Z, Motosugi U, et al. Diffusion-weighted imaging in optic neuritis. 30. Al-Shafai LS, Mikulis DJ. Diffusion MR imaging in a case of acute ischemic Can Assoc Radiol J. 2013;64:51–5. optic neuropathy. AJNR Am J Neuroradiol.2006;27(2):255–7. 8. Purohit BS, Vargas MI, Ailianou A, et al. Orbital tumours and tumour-like 31. Al-Zubidi N, Stevens S, Fung SH, et al. Diffusion-weighted imaging in pos- lesions: exploring the armamentarium of multiparametric imaging. terior ischemic optic neuropathy. Can J Ophthalmol. 2014;49(1):e21–e25. Insights Imaging. 2016;7:43–68. 32. Bender B, Heine C, et al. Diffusion restriction of the optic nerve in patients 9. Sapkota N, Shi X, et al. Two-dimensional single-shot diffusion-weighted with acute visual deficit. J Magn Reson Imaging. 2013;40:334–40. stimulated EPI with reduced FOV for ultrahigh-b radial diffusion-weighted 33. Yeom KW, Lober RM, Andre JB, et al. Prognostic role for diffusion- imaging of spinal cord. Magn Reson Med. 2017;77:2167–73. weighted imaging of pediatric optic pathway glioma. J Neurooncol. 10. Seeger A, Klose U, Bischof F, et al. Zoomed EPI DWI of acute spinal 2013;113(3):479–83. ischemia using a parallel transmission system. Clin Neuroradiol. 34. Barker GJ. Diffusion-weighted imaging of the spinal cord and optic nerve. 2016;26(2):177–82. J Neurol Sci. 2001;186(supp-S1):S45–9. 11. Mannelli L, Monti S, Corrias G, et al. Comparison of navigator triggering 35. Xu XQ, Liu J, Hu H, et al. Improve the image quality of orbital 3 T diffusion- reduced field of view and large field of view diffusion-weighted imaging weighted magnetic resonance imaging with readout-segmented echo- of the pancreas. J Comput Assist Tomogr. 2018;43(1):1. planar imaging. Clin Imaging. 2016;40(4):793–6. 12. Brendle C, Martirosian P, Schwenzer NF, et al. Diffusion-weighted imaging 36. Porter DA, et al. High resolution diffusion-weighted imaging using read- in the assessment of prostate cancer: comparison of zoomed imaging out-segmented echo-planar imaging, parallel imaging and a two-dimen- and conventional technique. Eur J Radiol. 2016;85(5):893–900. sional navigator-based reacquisition. Magn Reson Med. 2009;62:468–75. 13. Thierfelder KM, Scherr MK, Notohamiprodjo M, et al. Diffusion-weighted 37. Cohen-Adad J. High-resolution DWI in brain and spinal cord with syngo MRI of the prostate: advantages of Zoomed EPI with parallel- RESOLVE1. Siemens Magnetom-Clin Neurol. 2012;510:16–23. transmit-accelerated 2D-selective excitation imaging. Eur Radiol. 38. Seeger A, Batra M, Süsskind D, et al. Assessment of uveal melanomas 2014;24(12):3233–41. using advanced diffusion-weighted imaging techniques: value of 14. Dong H, Li Y, Li H, et al. Study of the reduced field-of-view diffusion- reduced field of view DWI (“zoomed DWI”) and readout-segmented DWI weighted imaging of the breast. Clin Breast Cancer. 2014;14(4):265–71. (RESOLVE). Acta Radiol. 2018;60:977–84. 15. Song M-H, Jin Y-F, et al. Application of whole-lesion intravoxel incoherent motion analysis using iZOOM DWI to differentiate malignant from benign Publisher’s Note thyroid nodules. Acta Radiol. 2018;60:1127–34. Springer Nature remains neutral with regard to jurisdictional claims in pub- 16. Yuan T, Wang J, et al. Comparison of field-of-view optimized and lished maps and institutional affiliations. constrained undistorted single-shot diffusion-weighted imaging and conventional diffusion-weighted imaging of optic nerve and chiasma at 3T. Neuroradiology. 2018;60(9):903–12. 17. Feng Z, Min X, Sah VK, et al. Comparison of field-of-view (FOV ) optimized and constrained undistorted single shot (FOCUS) with conventional DWI for the evaluation of prostate cancer. Clin Imaging. 2015;39:851–5. 18. Seeger A, Schulze M, Schuettauf F, et al. Advanced diffusion-weighted imaging in patients with optic neuritis deficit—value of reduced field of view DWI and readout-segmented DWI. Neuroradiol J. 2018;31:126–32. 19. Sha Y, Tian G, et al. Diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging, and two-dimensional navigator- based reacquisition in detecting acute optic neuritis. J Magn Reson Imaging. 2016;43:655–60. 20. Chang P, Nassirpour S, Henning AJ. Modeling real shim fields for very high degree (and order) B0 shimming of the human brain at 9.4 T. Magn Reson Med. 2018;79:529–40. 21. Porter DA, Heidemann RM. High resolution diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging and Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : a two-dimensional navigator-based reacquisition. Magn Reson Med. 2010;62(2):468–75. fast, convenient online submission 22. Hu L, Wei L, Wang S, et al. Better lesion conspicuity translates into thorough peer review by experienced researchers in your field improved prostate cancer detection: comparison of non-parallel- transmission-zoomed-DWI with conventional-DWI. Abdom Radiol (N Y ). rapid publication on acceptance 2021;46(12):5659–68. https:// doi. org/ 10. 1007/ s00261- 021- 03268-5. support for research data, including large and complex data types 23. Rosenkrantz AB, Chandarana H, Pfeuffer J, et al. Zoomed echo-planar • gold Open Access which fosters wider collaboration and increased citations imaging using parallel transmission: impact on image quality of diffusion-weighted imaging of the prostate at 3T. Abdom Imaging. maximum visibility for your research: over 100M website views per year 2015;40(1):120–6. 24. Kolbe S, Chapman C, Nguyen T, et al. Optic nerve diffusion changes and At BMC, research is always in progress. atrophy jointly predict visual dysfunction after optic neuritis. Neuroim- Learn more biomedcentral.com/submissions age. 2009;45(3):679–86.

Journal

BMC Medical Imaging – Springer Journals

Published: May 24, 2022

Keywords: Optic nerves; Diffusion-weighted imaging; Zoomed EPI; RESOLVE

There are no references for this article.