Readout-segmented echo-planar diffusion-weighted imaging in the assessment of orbital tumors: comparison with conventional single-shot echo-planar imaging in image quality and diagnostic performance

Abstract

Background

Readout-segmented echo-planar imaging (RS-EPI) could improve the imaging quality of diffusion-weighted imaging (DWI) in various organs. However, whether it could improve the imaging quality and diagnostic performance for the patients with orbital tumors is still unknown.

Purpose

To compare the image quality and diagnostic performance of RS-EPI DWI with that of conventional single-shot EPI (SS-EPI) DWI in patients with orbital tumors.

Material and Methods

SS-EPI and RS-EPI DW images of 32 patients with pathologically diagnosed orbital tumors were retrospectively analyzed. Qualitative imaging parameters (imaging sharpness, geometric distortion, ghosting artifacts, and overall imaging quality) and quantitative imaging parameters (apparent diffusion coefficient [ADC], signal-to-noise ratio [SNR], contrast, and contrast-to-noise ratio [CNR]) were assessed by two independent radiologists, and compared between SS-EPI and RS-EPI DWI. Receiver operating characteristic curves were used to determine the diagnostic value of ADC in differentiating malignant from benign orbital tumors.

Results

RS-EPI DW imaging produced less geometric distortion and ghosting artifacts, and better imaging sharpness and overall imaging quality than SS-EPI DWI (for all, P < 0.001). Meanwhile, RS-EPI DWI produced significantly lower SNR (P < 0.001) and ADC (P < 0.001), and higher contrast (P < 0.001) than SS-EPI DWI, while producing no difference in CNR (P = 0.137). There was no significant difference on the diagnostic performance between SS-EPI and RS-EPI DWI, when using ADC as the differentiating index (P = 0.529).

Conclusion

Compared with SS-EPI, RS-EPI DWI provided significantly better imaging quality and comparable diagnostic performance in differentiating malignant from benign orbital tumors.

Keywords

Orbit magnetic resonance imaging (MRI)diffusion-weighted imaging readout-segmented echo-planar imaging single-shot echo-planar imaging

Introduction

Diffusion-weighted imaging (DWI), which enables measurement of apparent diffusion coefficient (ADC), is considered an essential magnetic resonance imaging (MRI) method for diagnosing orbital disease (1 –3). DWI has shown potential as an imaging biomarker for differentiating benign and malignant orbital lesions, and in assessing therapeutic response (4 –11). In current clinical practice, single-shot echo-planar imaging (SS-EPI) is used most widely due to its resistance to patient motion and scanning speed (12). However, SS-EPI can result in geometric distortions, image blurring, ghosting artifacts, and problems related to fat suppression. These detrimental effects would become more severe at high-field MR strengths (13). Thus, the diagnostic ability of DWI based on SS-EPI would be limited.

Standard SS-EPI DWI is prone to distortion, primarily because of its long readout time and low bandwidth in the phase-encoding direction (14). Thus, previous investigators have developed several methods to address this issue, including half-Fourier acquisition single-shot turbo spin-echo imaging (HASTE), single-shot fast spin-echo, line-scan imaging, a reduced field-of-view (FOV) approach, and interleaved and parallel EPI approaches (14). These methods may yield high-quality images, but are not widely used owing to long acquisition times and the increased specific absorption rate for clinical resolution.

DWI based on readout segmented EPI (RS-EPI), with two-dimensional (2D) navigator echoes for reduced sensitivity to motion, has been suggested as an alternative imaging approach (12 –22). In setting up RS-EPI DWI, the k space is divided into multiple segments along the direction of the readout. This permits the usage of shortened echo spacing in each segment, and can reduce geometric distortion and susceptibility artifacts by accelerating the k-space traversal along the direction of the readout (22). These superiorities of RS-EPI DWI have been found in the imaging scans of various organs at 3.0 T MRI, including pediatric and adult brains (13,15 –17), breasts (19), kidneys (20,22), and the head and neck regions (21). In addition, one prior study has validated the feasibility of RS-EPI in orbital DWI, and reported that RS-EPI could improve the image quality of orbital DWI in healthy participants (23). However, to the best of our knowledge, no study has compared the lesion visualization and diagnostic performance of DWI based on standard SS-EPI and RS-EPI in patients with orbital tumors at 3.0 T.

Therefore, the purpose of our study is to evaluate the clinical utility of orbital RS-EPI DWI, and to compare its imaging quality and diagnostic performance with that of SS-EPI DWI in patients with orbital tumors at 3.0 T.

Material and Methods

Patients

This study protocol was approved by the institutional review board of our hospital; written informed consent was waived due to the retrospective nature of our study. The following inclusion criteria were applied to constitute the final group of 32 patients with orbital tumors (17 men, 15 women; mean age, 53 ± 13 years; age range, 25–88 years) in our study: (i) both SS-EPI and RS-EPI DWI were performed for pre-treatment MRI evaluation; (ii) the largest diameter of the lesion exceeded 1 cm; (iii) the image quality was adequate for further analysis; and (iv) the diagnosis was made based on the pathological examination

The group of 32 patients contained 21 patients with benign tumors (9 men, 12 women; mean age, 50 ± 10 years; age range, 25–66 years) and 11 patients with malignant tumors (8 men, 3 women; mean age, 58 ±16 years; age range, 28–88 years). Twenty-one benign orbital lesions comprised cavernous malformation (n = 10), pleomorphic adenoma of the lacrimal gland (n = 5), schwannoma (n = 3) and inflammatory pseudotumor (n = 2), and solitary fibrous tumor (n = 1). Eleven malignant orbital lesions comprised lymphoma (n = 10) and metastasis (n = 1).

MRI

MRI was performed with a 3.0-T MR scanner (Magnetom Verio; Siemens Healthcare, Erlangen, Germany) with a 12-channel head coil, with patients resting in the supine position. Conventional MRI was acquired using the following sequences: axial T1-weighted (T1W) imaging (repetition time/echo time [TR)/TE], 600/10 ms); axial, coronal, and sagittal T2-weighted (T2W) imaging (TR/TE, 4700/79 ms) with fat saturation; axial, coronal, and sagittal post-contrast T1W imaging with fat saturation (TR/TE, 600/10 ms).

Before contrast media was used, SS-EPI and RS-EPI DWI were performed with comparable imaging parameters (Table 1). To match the acquisition time of the RS-EPI DWI, the average of SS-EPI DWI was increased to 5, and the average of RS-EPI DWI was kept at 1.

Table 1.

Sequence parameters for SS-EPI and RS-EPI.

Sequence parameter	SS-EPI	RS-EPI
Diffusion direction	Three-direction trace	Three-direction trace
Diffusion schema	Stejskal-Tanner	Stejskal-Tanner
b value (s/mm²)	0, 1000	0, 1000
Fat suppression	Frequency selective	Frequency selective
Repetition time (ms)	4000	4000
Echo time (ms)	85	69
Field of view (mm)	220	220
Matrix	130 × 130	130 × 130
Numbers of sections	9	9
Section thickness (mm)	4	4
Intersection gap (%)	30	30
Phase-encoding direction	Anteroposterior	Anteroposterior
Echo-spacing (m)	1.07	0.4
Readout segments (n)	1	5
Signals acquired (n)	5	1
Acquisition time (min:s)	1:34	1:50

Qualitative comparisons of image quality

The images were evaluated focusing on the following four characteristics: image sharpness, geometric distortion, ghosting artifacts, and overall image quality. The detailed imaging criteria and the corresponding scores are listed in Table 2.

Table 2.

Criteria for qualitative comparison of image quality in DWI in patients with orbital tumors.

Image sharpness (clear depiction of lesion edge)

1: Clearly visible

2: Partially indefinitely visible

3: Completely indefinitely visible

Geometric distortion (distortion in size, profile, or orientation)

1: No distortion

2: Moderate distortion

3: Marked distortion

Ghosting artifact (the interface of the orbital apex)

1: No artifact

2: Faint artifact, no impact on imaging interpretation

3: Obvious artifact, impact on imaging interpretation

Overall image quality

1: Non-diagnostic

2: Poor

3: Satisfactory

4: Good

4: Excellent

Quantitative comparisons of image quality

SNR was defined as the ratio between mean signal amplitude inside the region of interest (S_ROI) and standard deviation of the background noise (S_BG) (SNR = S_ROI/Ơ_BG) (24). Contrast was defined as the ratio between the mean signal intensity of the lesion (S_ROI) and that of the brainstem on DW images (S_B) (contrast = S_ROI/S_B) (24). CNR was defined as the difference between S_ROI and S_B divided by the standard deviation in the lesion ROI (Ơ_ROI) and brainstem ROI (Ơ_B), as expressed by the following formula (24):

CNR = \frac{(S_{ROI} - S_{E})}{\sqrt{O_{{ROI}^{2}} + O_{E^{2}}}}

During the measurements of CNR, contrast, and SNR, the slice on which the orbital mass showed the largest diameter was chosen. Three circular ROIs were retrospectively placed on the mass (yellow), brainstem (green), and background (red) (Fig. 1a) on the DW image (b = 1000 s/mm²). Then, the CNR, contrast, and SNR were calculated manually according to the abovementioned formula.

Fig. 1.

ROI placements for the quantitative measurements. For the measurements of SNR, contrast, and CNR, three circular ROIs were placed on the mass (yellow), brainstem (green), and background (red), retrospectively (a). Then, they were calculated manually according to the corresponding formula. For the measurement of ADC value, ROIs were placed on each slice of the whole tumor (b). Then, the ADC value was calculated automatically by software.

In terms of the measurement of ADC, all imaging data were post-processed offline using in-house software (FireVoxel; CAI²R; New York University, NY, USA) (25). Whole-tumor ROIs were drawn on every slice encompassing the lesion (Fig. 1b), and ADC was calculated using the following formula: ADC = −ln(S_b/S₀)/b, where b represents the diffusion sensitivity coefficients and S_b and S₀ represent the corresponding signal values of the given ROIs.

Two independent radiologists (reader 1, with 15 years of experience; reader 2, with 5 years of experience), who were blinded to the study design and pathological results, retrospectively performed the qualitative and quantitative measurements. The average of the two measurements was adopted for analysis.

Statistical analysis

Continuous variables were expressed as mean ± standard deviation (SD), and the normality was tested using the Kolmogorov–Smirnov test. The differences of qualitative and quantitative parameters between SS-EPI and RS-EPI DW images were compared using paired t-tests. Receiver operating characteristic (ROC) curves were generated to evaluate the diagnostic ability of ADC values that were obtained from SS-EPI and RS-EPI DW images in differentiating orbital benign and malignant tumors, and then the area under curves (AUC) were compared using the method of Delong et al. (26).

The inter-reader variability of qualitative and quantitative metrics was assessed using Kappa analyses and intra-class correlation (ICC) coefficient, respectively (27). Bland–Altman plots were used to assess agreements in ADC values (28). The kappa and ICC coefficients were in the range of 0–1.00, and they were interpreted as follows: < 0.40, poor; 0.41–0.60, moderate; 0.61–0.80, good; r ≥ 0.81, excellent. Statistical analyses were performed using SPSS software (SPSS, version 19.0; SPSS, Chicago, IL, USA) and MedCalc software (MedCalc, version 9.0; MedCalc Software, Mariakierke, Belgium). A P value of 0.05 and less indicated a significant difference.

Results

Qualitative analysis

During the qualitative comparison of SS-EPI and RS-EPI DW images, the detailed inter-reader kappa values were 0.783 and 0.799 for image sharpness, 0.821 and 0.845 for geometric distortion, 0.819 and 0.836 for ghosting artifacts, 0.743 and 0.812 for overall imaging quality of SS-EPI and RS-EPI (Table 3). These results indicated the good to excellent agreement between the two readers.

Table 3.

Qualitative and quantitative parameters between RS-EPI and SS-EPI.

Parameters	SS-EPI	RS-EPI	P	Inter-reader variability
Parameters	SS-EPI	RS-EPI	P	SS-EPI	RS-EPI
Qualitative parameters				k
Imaging sharpness	2.313 ± 0.454	1.141 ± 0.261	<0.001	0.783	0.799
Geometric distortion	2.609 ± 0.416	1.203 ± 0.333	<0.001	0.821	0.845
Ghosting artifact	2.672 ± 0.373	1.344 ± 0.410	<0.001	0.819	0.836
Overall imaging quality	2.750 ± 0.751	4.359 ± 0.571	<0.001	0.743	0.812
Quantitative parameters				ICC
ADC	1.032 ± 0.336	0.977 ± 0.319	<0.001	0.991	0.988
SNR	134.870 ± 64.682	51.350 ± 19.241	<0.001	0.930	0.910
Contrast	0.875 ± 0.374	1.278 ± 0.615	<0.001	0.894	0.885
CNR	6.159 ± 4.928	4.650 ± 3.973	0.137	0.825	0.891

Except for the P values, data are reported as mean ± standard deviation. Qualitative and quantitative measurements are based on all orbital tumors. Unit for ADC value is × 10^–3 mm²/s.

ADC, apparent diffusion coefficient; CNR, contrast-to-noise ratio; ICC, intra-class correlation coefficient; RS-EPI, readout-segmented echo-planar imaging; SNR, signal-to-noise ratio; SS-EPI, single-shot echo-planar imaging.

Table 1 summarizes the average scores of the qualitative assessments of SS-EPI and RS-EPI. RS-EPI was superior to SS-EPI DW images regarding image sharpness (2.313 ± 0.454 versus 1.141 ± 0.261, P < 0.001), geometric distortion (2.609 ± 0.416 versus 1.203 ±0.333, P < 0.001), ghosting artifacts (2.672 ±0.373 versus 1.344 ± 0.410, P < 0.001), and overall imaging quality (2.750 ± 0.751 versus 4.359 ± 0.571, P < 0.001) (Fig. 2). Representative SS-EPI and RS-EPI DW images of the patients with orbital tumors are shown in Figs. 3 and 4.

Fig. 2.

Bar graph showing the comparison between SS-EPI and RS-EPI diffusion weighted images in terms of four qualitative imaging parameters.

Fig. 3.

T2W image (a), SS-EPI DW image (b), and RS-EPI DW image (c) of a 48-year-old woman with pleomorphic adenoma of the lacrimal gland. A marked geometric distortion of the vitreous body was seen on the SS-EPI DW image, while these were significantly reduced on the RS-EPI DW image.

Fig. 4.

T2W image (a), SS-EPI DW image (b), and RS-EPI DW image (c) of a 72-year-old man with orbital lymphoma. A marked distortion artifact that located behind the mass was seen on the SS-EPI DW image, while it was significantly reduced on the RS-EPI DW image.

Quantitative analysis

Detailed inter-reader ICC values for the measurements of quantitative parameters are listed in Table 3. The ICC values were 0.930 and 0.910 for SNR, 0.894 and 0.885 for contrast, and 0.825 and 0.891 for CNR of SS-EPI and RS-EPI DW images. These results indicated the excellent agreement between the two readers.

The comparisons of quantitative image parameters between SS-EPI and RS-EPI are listed in Table 3. SNR, contrast, and CNR were 134.870 ± 64.682, 0.875 ±0.374, and 6.159 ± 4.928 for SS-EPI, respectively, and 51.350 ± 19.241, 1.278 ± 0.615, and 4.650 ± 3.973 for RS-EPI, respectively. RS-EPI was superior to SS-EPI in SNR (P < 0.001) and contrast (P < 0.001), but not in CNR (P = 0.137).

Comparison of ADC value and diagnostic ability

Excellent inter-reader agreements were also obtained during the measurements of ADC values from SS-EPI or RS-EPI DW images (ICC, 0.991 and 0.988) (Table 3). Representative Bland–Altman plots reflecting the agreements of the measurements of ADC value from SS-EPI or RS-EPI DW images are shown in Fig. 5. The mean ADC values of all the orbital tumors on the RS-EPI DW images were significantly lower than those on the SS-EPI DW images (P < 0.001).

Fig. 5.

Bland–Altman plots showing the reproducibility of mean ADC derived from SS-EPI and RS-EPI DW images of orbital tumors. Difference of mean ADC value between two readers (y axis) was plotted against mean ADC value of two readers (x axis), with mean absolute difference (bias) (solid line) and 95% confidence interval of the mean difference (limits of agreement) (dashed lines). The plots in the figure showed the absolute ADC values from reader 2, while setting the ADC values from reader 1 as the standard.

Based on both SS-EPI and RS-EPI DWI, the mean ADC value of malignant tumors was significantly lower than that of benign mimics (SS-EPI, 0.715 ± 0.365 versus 1.199 ± 0.157, P < 0.001; RS-EPI, 0.666 ± 0.338 versus 1.140 ± 0.138, P < 0.001) (Fig. 6). Table 4 summarizes the diagnostic performance of ADC value from SS-EPI or RS-EPI DW images in differentiating orbital malignant from benign tumors. There was no significant difference in the AUC for the mean ADC value (AUC, SS-EPI versus RS-EPI, 0.887 versus 0.896, P = 0.529) (Fig. 7).

Fig. 6.

Bar graph showing the comparisons of ADC values between malignant and benign group.

Table 4.

Diagnostic ability of ADC value obtained from RS-EPI and SS-EPI for differentiating orbital benign and malignant tumors.

Parameters	Cutoff value	AUC	Sensitivity (100%)	Specificity (100%)
ADC (SS-EPI)	0.727	0.887 (0.726–0.971)	81.8 (48.2–97.7)	100 (83.9–100)
ADC (RS-EPI)	0.693	0.896 (0.737–0.976)	81.8 (48.2–97.7)	100 (83.9–100)

Unit for ADC value is 10⁻³ mm²/s. Data in parentheses indicates 95% confidence interval.

ADC, apparent diffusion coefficient; AUC, area under curve; RS-EPI, readout-segmented echo-planar imaging; SS-EPI, single-shot echo-planar imaging.

Fig. 7.

ROC curves of using mean ADC value derived from SS-EPI and RS-EPI DW images to differentiate benign and malignant orbital tumors.

Discussion

In our study, we demonstrated higher image quality of orbital DWI using RS-EPI, compared with the standard SS-EPI technique. In addition, no significant difference was found on the diagnostic performance of ADC value from SS-EPI and RS-EPI DW images for differentiating orbital malignant from benign tumors. RS-EPI technique significantly improved the image quality, while maintaining comparable diagnostic performance. Our study indicated RS-EPI might be a promising approach for quantitating orbital DWI.

In terms of the qualitative imaging assessment, RS-EPI was found to be superior to SS-EPI DWI in reducing geometric distortion and ghosting artifacts, while increasing image sharpness and overall image quality, which was consistent with previous studies (13 –24). In RS-EPI DWI, the k space was divided into several segments along the direction of the readout. This permitted the use of shortened echo spacing in each segment, and then the reduction of T2* blurring, geometric distortion, and susceptibility artifacts by accelerating the k-space traversal along the direction of readout (22,29,30). Moreover, a navigator-based reacquisition technique was performed for phase correction in RS-EPI DWI, which can provide potential correction for shot-to-shot phase artifacts (22). Considering the abovementioned sequence characteristic, it was not surprising that RS-EPI could significantly improve the overall image quality of orbital DWI.

Controversial results have always existed in comparisons of SNR between SS-EPI and RS-EPI. Previously, Wu et al. found that RS-EPI had a higher SNR in renal DW images than SS-EPI did in an animal study (22). However, Bogner et al. insisted that the SNR of RS-EPI DWI of the breast was lower than that of SS-EPI (24). Our study agreed with the latter study. This can be explained by the more efficient k-space coverage of SS-EPI and higher actual resolution of RS-EPI (24). In addition, we added the average of the SS-EPI to 5, and therefore it was not surprising that the SNR of SS-EPI was higher than that of RS-EPI. Regarding SNR, we considered that it was difficult to conclude which sequence would have higher SNR, because SNR can be influenced by several factors, such as the optical TE, the effect of PSF, and the T2 value of the tissue measured (23). SNR depends on the specific protocol actually used to a large extent. In this study, no significant difference was found in the CNR between RS-EPI and SS-EPI, which was consistent with previous studies (24,29). Moreover, similar to previous studies, the contrast of orbital tumors on DW images was significantly better for RS-EPI than for SS-EPI, thus enabling better visual discrimination of orbital tumors on RS-EPI DWI (24).

As to the ADC value, controversial results have always existed among previous studies. Koyasu et al. found that there was no significant difference in the ADC value of parotid gland lesions between RS-EPI and SS-EPI DW images (21). However, Wisner et al. (31) and Zhao et al. (29) found that the mean ADC values derived from RS-EPI were lower than those from SS-EPI DWI. Our study results were in accordance with the latter two studies. As to the potential reason, Wisner et al. indicated that T2* blurring in SS-EPI DW images would make the ADC measurement be influenced by adjacent normal tissue with high ADC value (31). While Zhao et al. proposed that the distortion and artifacts might be a more reasonable explanation, because no adjacent normal tissue with high ADC value could be found in the sinonasal region (29). The distortion and artifacts which were more commonly seen in SS-EPI DW images, could result in the inhomogeneous ADC maps, and consequently make the measured ADC value increased. In our opinion, we give more support to the latter explanation. One prior study found no significant difference in the ADC value of the brainstem in the healthy volunteers between two DW images (23), therefore the former explanation does not seem reasonable. The difference might mainly result from the aforementioned distortion and artifacts. Therefore, the ADC value measured from RS-EPI DW images might be more accurate for the quantitation of Brownian motion of water molecules in orbital tumors.

In our study, we found the mean ADC value of malignant orbital tumors was significantly lower than that of benign mimics, and an ADC value of 0.727 × 10⁻³ mm²/s from SS-EPI DWI was optimal for the differentiation (sensitivity 81.8%, specificity 100%;). The cutoff value, sensitivity, and specificity were different from Razek et al. (cutoff value, 1.15 × 10⁻³ mm²/s; sensitivity 95%, specificity 91%) (8) and Sepahdari et al. (cutoff value, 1.0 ×10⁻³ mm²/s; sensitivity 63%, specificity 84%) (9). In our opinion, this controversy might be associated with the different magnetic field strength, MR acquisition parameters, and patients’ compositions (3).

Previously, Boner et al. found that, on the basis of an ADC threshold of 1.25 × 10⁻³ mm²/s, RS-EPI demonstrated a higher diagnostic accuracy for differentiating benign and malignant breast lesions than SS-EPI DW images (24). However, we did not find superior diagnostic ability for RS-EPI DW image in the present study. Limited sample size and monotonous pathological composition might be the potential reason for the non-superiority. Besides that, the ADC cutoff value derive from RS-EPI was different from that from SS-EPI. It implied that the ADC cutoff value derive from SS-EPI DW imaging could not be directly used in RS-EPI DWI related studies. Further study with larger sample size and more pathological compositions could help to clarify to the cutoff ADC value from RS-EPI DWI for differentiation, and provide more definitive evidence of the effect of RS-EPI DWI on diagnostic performance.

Our study had several limitations. First, the TE of the SS-EPI was slightly longer than that of the RS-EPI (85 versus 69 ms). This fact would benefit the RS-EPI DW image (14). Second, the ADC values derived from two DWI techniques were not correlated with histopathologic examination. Further studies performing this correlation analysis would help to clarify the value of ADC from RS-EPI DW images for accurately quantitating the Brownian motion of water molecules in orbital tumors. Third, the pathological composition of malignant group is very monotonous. Most of the cases were lymphoma patients. Therefore, further comparison between lymphoma and other malignancies could not been performed. Last, this study was performed at a single center with limited sample size. Further multicenter studies with a larger number of patients are needed to confirm our study results.

In conclusion, our study showed that RS-EPI can significantly improve the image quality of orbital DWI by reducing the susceptibility artifacts, distortion, and imaging blurring. RS-EPI DWI could be a robust tool for delineating the contours of tumors. Compared to SS-EPI, RS-EPI DWI showed comparable diagnostic performance in differentiating benign and malignant orbital tumors. Our study results confirm the feasibility of further usage of RS-EPI for orbital DWI in a clinical setting.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Tailor

Gupta

Dalley

et al.

Orbital neoplasms in adults: clinical, radiologic, and pathologic review. Radiographics 2013; 33: 1739–1758.

Asbach

Siebert

et al.

Characterization of orbital masses by multiparametric MRI. Eur J Radiol 2016; 85: 324–336.

Haradome

Usui

et al.

Orbital lymphoproliferative disorders (OLPDs): value of MR imaging for differentiating orbital lymphoma from benign OPLDs. Am J Neuroradiol 2014; 35: 1976–1982.

Sepahdari

Kapur

Aakalu

et al.

Diffusion-weighted imaging of malignant ocular masses: initial results and directions for further study. Am J Neuroradiol 2012; 33: 314–319.

et al.

Orbital indeterminate lesions in adults: combined magnetic resonance morphometry and histogram analysis of apparent diffusion coefficient maps for predicting malignancy. Acad Radiol 2016; 23: 200–208.

et al.

Utility of histogram analysis of ADC maps for differentiating orbital tumors. Diagn Interv Radiol 2016; 22: 161–167.

Liu

et al.

Benign and malignant orbital lymphoproliferative disorders: Differentiating using multiparametric MRI at 3.0T. J Magn Reson Imaging 2017; 45: 167–176.

Razek

Elkhamary

Mousa

. Differentiation between benign and malignant orbital tumors at 3-T diffusion MR imaging. Neuroradiology 2011; 53: 517–522.

Sepahdari

Aakalu

Setabutr

et al.

Indeterminate orbital masses: restricted diffusion at MR imaging with echo-planar diffusion-weighted imaging predicts malignancy. Radiology 2010; 256: 554–564.

10.

Politi

Forghani

Godi

et al.

Ocular adnexal lymphoma: diffusion-weighted MR imaging for differential diagnosis and therapeutic monitoring. Radiology 2010; 256: 565–574.

11.

Guo

Liu

Yan

et al.

Prediction of the response of ocular adnexal lymphoma to chemotherapy using combined pretreatment dynamic contrast-enhanced and diffusion-weighted MRI. Acad Radiol 2016; 57: 1490–1496.

12.

Sun

Chai

et al.

Diffusion-weighted imaging-guided mr spectroscopy in breast lesions using readout-segmented echo-planar imaging. Eur Radiol 2016; 26: 1565–1574.

13.

Byeon

Kim

Cho

. Readout-segmented echo-planar imaging in diffusion-weighted MR imaging of acute infarction of the brainstem and posterior fossa: comparison of single-shot echo-planar diffusion-weighted sequences. Clin Imaging 2015; 39: 765–769.

14.

Wang

Deng

et al.

Feasibility study of 3-T DWI of the prostate: readout-segmented versus single-shot echo-planar imaging. Am J Roentgenol 2015; 205: 70–76.

15.

Yeom

Holdsworth

Van

et al.

Comparison of readout-segmented echo-planar imaging (EPI) and single-shot EPI in clinical application of diffusion-weighted imaging of the pediatric brain. Am J Roentgenol 2013; 200: W437–443.

16.

Holdsworth

Yeom

Skare

et al.

Clinical application of readout-segmented- echo-planar imaging for diffusion-weighted imaging in pediatric brain. Am J Neuroradiol 2011; 32: 1274–1279.

17.

Morelli

Porter

et al.

Clinical evaluation of single-shot and readout-segmented diffusion-weighted imaging in stroke patients at 3 T. Acta Radiol 2013; 54: 299–306.

18.

Wan

Sha

Zhang

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: 655–660.

19.

Park

Shin

et al.

Comparison of readout segmented echo planar imaging (EPI) and EPI with reduced field-of-View diffusion-weighted imaging at 3t in patients with breast cancer. J Magn Reson Imaging 2015; 42: 1679–1688.

20.

Friedli

Crowe

Viallon

et al.

Improvement of renal diffusion-weighted magnetic resonance imaging with readout-segmented echo-planar imaging at 3T. Magn Reson Imaging 2015; 33: 701–708.

21.

Koyasu

Iima

Umeoka

et al.

The clinical utility of reduced-distortion readout-segmented echo-planar imaging in the head and neck region: initial experience. Eur Radiol 2014; 24: 3088–3096.

22.

Wang

Zhang

et al.

Readout-segmented echo-planar imaging in diffusion-weighted imaging of the kidney: comparison with single-shot echo-planar imaging in image quality. Abdom Radiol (NY) 2016; 41: 100–108.

23.

Liu

et al.

Improve the image quality of orbital 3 T diffusion-weighted magnetic resonance imaging with readout-segmented echo-planar imaging. Clin Imaging 2016; 40: 793–796.

24.

Bogner

Pinker-Domenig

Bickel

et al.

Readout-segmented echo-planar imaging improves the diagnostic performance of diffusion-weighted MR breast examinations at 3.0 T. Radiology 2012; 263: 64–76.

25.

Wang

et al.

DWI-associated entire-tumor histogram analysis for the differentiation of low-grade prostate cancer from intermediate-high-grade prostate cancer. Abdom Imaging 2015; 40: 3214–3221.

26.

DeLong

Clarke-Pearson

. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988; 44: 837–845.

27.

Cohen

. Weighted kappa: nominal scale agreement with provision for scaled disagreement or partial credit. Psychol Bull 1968; 70: 213–220.

28.

Bland

Altman

. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–310.

29.

Zhao

Liu

Sha

et al.

Readout-segmented echo-planar imaging in the evaluation of sinonasal lesions: A comprehensive comparison of image quality in single-shot echo-planar imaging. Magn Reson Imaging 2016; 34: 166–172.

30.

Kim

Kang

et al.

Readout-segmented echo-planar imaging in diffusion-weighted MR imaging in breast cancer: comparison with single-shot echo-planar imaging in image quality. Korean J Radiol 2014; 15: 403–410.

31.

Wisner

Rogers

Deshpande

et al.

High-resolution diffusion-weighted imaging for the separation of benign from malignant BI-RADS 4/5 lesions found on breast MRI at 3T. J Magn Reson Imaging 2014; 40: 674–681.