Abstract
Objectives
To explore the utility of whole-lesion apparent diffusion coefficient (ADC) histogram analysis for differentiating parotid gland tumors following readout-segmented diffusion-weighted imaging (RESOLVE).
Methods
80 patients (40 with pleomorphic adenomas, 14 with Warthin tumors, and 26 with malignant parotid gland tumors) who underwent routine head-and-neck MRI and RESOLVE examinations, were retrospectively evaluated. RESOLVE data were acquired from a MAGNETOM Skyra 3T MR system. Eleven whole-lesion histogram parameters derived from histogram analysis (ADC_mean, ADC_minimum, ADC_maximum, ADC_1th, ADC_10th, ADC_50th, ADC_90th, ADC_99th, skewness, variance and kurtosis) were calculated for each patient using MaZda. Receiver operating characteristic (ROC) curve analysis was used to assess the diagnostic performance of the ADC for distinguishing among the three groups.
Results
In total, nine parameters (ADC_minimum, ADC_maximum, ADC_mean, ADC_10th, ADC_50th, ADC_90th, ADC_99th, variance, skewness) were statistically significant (all p < 0.05) for all three groups, in the comparison of pleomorphic adenomas to Warthin tumors; the ADC_mean, ADC_50th, and skewness revealed high diagnostic efficiency with areas under the receiver operating characteristic curve of 0.976, 0.970, and 0.970, respectively. In the comparison of pleomorphic adenomas to malignant parotid gland tumors, these nine parameters were also found to be statistically different (all p < 0.05); the ADC_mean, ADC_10th and ADC_50th revealed high diagnostic efficiency with area under the curve of 0.851, 0.866, and 0.841, respectively. However, in the comparison of Warthin tumors to malignant parotid gland tumors, only three parameters (ADC_mean, ADC_50th, skewness) were statistically significant (all p < 0.05).
Conclusions
Whole-lesion ADC histograms are effective in differentiating common parotid gland tumors.
Keywords: apparent diffusion coefficient, parotid Gland, histogram, tumor
Introduction
About 80% of salivary glands tumors originate from the parotid gland. Parotid gland tumors include a wide variety of histopathological types, the majority of which are benign. The preoperative analysis of parotid gland tumors plays a critical role in surgical planning because the surgical approach and prognosis differ significantly depending on the diagnosis.1 However, fine-needle aspiration cytology is invasive and the results are not always conclusive.2,3
Traditional MRI provides a good delineation of the lesion and the relation to anatomical structures; diffusion-weighted imaging (DWI) is also becoming an important tool for parotid gland tumor characterization. Previous researchers have reported that combining DWI with apparent diffusion coefficient (ADC) can improve the diagnostic accuracy of parotid gland tumors.4,5However, mean ADC values are often calculated based on a single region of interest (ROI) drawn over the largest area of the tumor or over the solid part of the tumor,6 this freehand approach is not always significantly sensitive to small changes.
With progressive advances in both high-resolution MRI and signal processing methods, histogram analyses of MRI are increasingly being used in the characterization of tumors.7 The histogram of an image refers to a histogram of the pixel intensity values, which shows the number of pixels in the whole image having the same intensity. Histogram parameters, including mean value, standard deviation, maximum, minimum, kurtosis, skewness, and percentiles (ADC_1th, ADC_10th, ADC_50th, ADC_90th, ADC_99th), are used to characterize the distributions of tumor biomarkers in a quantitative manner. Histogram analysis based on ADC mapping is helpful in the differentiation of benign tumors from malignant tumors originating from different organs, such as brain and pancreas.7,8 This technique can also be used to predict treatment response.9 In addition, readout-segmented echoplanar imaging (RS-EPI) utilizes multiple readout segments of k space along the readout direction, which reduces the effect of susceptibility and T2* blurring.10 The combination of RS-EPI and parallel imaging provides a further reduction in susceptibility and blurring artifacts by shortening the EPI echo-train length and decreasing the echo-spacing. RS-EPI can offer higher image quality in the head and neck area with a much shorter echo-spacing than single-shot echoplanar imaging (SS-EPI).11,12 Siemens Healthcare provided the prototype RS-EPI DWI sequence (RESOLVE) based on the RS-EPI technique. The implementation of RESOLVE relys on two-dimensional navigator correction, which provides a robust correction for motion-induced phase artifact and controls the real-time reacquisition of unusable data that cannot be corrected. The images acquired from RESOLVE produce achieve higher signal–noise ratios and resolution than those from SS-EPI, and the technique has been applied to many organs such as the pelvis, the breast and the kindey.13–15 However, to our knowledge, there are few reports of histogram analysis using RESOLVE to characterize parotid gland tumors.16
The purpose of this study was to explore the utility of whole-lesion ADC histogram analysis for differentiating benign from malignant parotid gland tumors following RESOLVE imaging.
Methods and materials
Patients
80 patients were histopathologically diagnosed with parotid gland tumors from June 2013 to January 2018 in this retrospective study. All patients underwent conventional MRI, contrast enhancement MRI, and RESOLVE at the first Affiliated Hospital of Zhengzhou University (Zhengzhou, P. R. China) . These patients had not undergone previous biopsies, surgeries, or treatments. The final diagnoses were made histologically using either surgery or biopsy. Eleven subjects were excluded from the study for one of three reasons: having a maximum lesion diameter <5 mm (n = 4), having a cystic lesion (n = 4), or because of motion artifacts (n = 3). In the end, 69 patients with parotid gland masses were enrolled in this study.
MR examinations
MRI examinations were performed on a 3T MR scanner (Magnetom Skyra, Siemens Healthcare, Erlangen, Germany) with an integrated 20-channel head and neck coil. Pre- and post-contrast transverse turbo spin echo (TSE) T1 weighted imaging (T1WI) with fat suppression (repetition time/echo time, 717/11 ms; 230 × 230 mm field of view, 224 × 320 matrix; 4 mm slice thickness; voxel size reconstructed to 0.7 × 0.7×4.0 mm; acquisition time, 1 min 6 s) and TSE T2 weighted imaging (T2WI) with fat suppression (repetition time/echo time, 4300/111 ms; 230 × 230 mm field of view; 224 × 320 matrix; 4 mm slice thickness; voxel size reconstructed to 0.7 × 0.7×4.0 mm ; acquisition time,1min 13 s) were obtained. Thereafter, RESOLVE (repetition time/echo time, 3900/64 ms; five readout segments; echo spacing, 0.36 ms; b-values = 0 and 1000 s/mm2; 240 × 240 mm field of view; 160 × 160 matrix; 4 mm slice thickness; voxel size, 1.5 × 1.5×4.0 mm; parallel acceleration factor of 2; acquisition time, 1 min 47 s) was obtained at the same section position as the transverse T1WI prior to the injection of contrast. Gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA, Magnevist, Schering, Berlin, Germany) was administered intravenously at a rate of 1.5 ml s−1 (total dose, 0.1 mmol per kg of body weight) using a power injector (MEDRAD Spectris Solaris EP, Bayer, Germany), followed by a 20 ml saline flush.
Image analysis
ADC maps were digitally transferred to MaZda software (v. 4.6, http:// www.eletel.p.lodz.pl/mazda/), and histogram analyses were conducted on a voxel-by-voxel basis. The ROI was manually drawn by two experienced radiologists around the whole tumor margin on each slice of ADC map, and then a histogram was made for each ROI (Figures 1–3). The pre- and post-contrast MR images and DW images were used as anatomic references to ensure that the entire lesion was chosen. DWI is a practical technique based on the random motion of water protons in biologic tissues. The following parameters were derived from the ADC histograms: the ADC_mean; ADC_minimum; ADC_maximum; ADC_1th; ADC_10th; ADC_50th; ADC_90th; ADC_99th; skewness; variance, and kurtosis.
Figure 1. .
ROI measurement and histogram distributions of pleomorphic adenomas measures of ADC map in MaZda. The abscissa represents different gray values within the ROI, and the ordinate represents the occurrence frequency of each gray value. ADC, apparent diffusion coefficient; ROI, region of interest.
Figure 2. .
ROI measurement and histogram distributions of Warthin tumors measures of ADC map in MaZda. The abscissa represents different gray values within the ROI, and the ordinate represents the occurrence frequency of each gray value. ADC, apparent diffusion coefficient; ROI, region of interest.
Figure 3a–c. .
ROI measurement and histogram distributions of adenoid cystic carcinoma measures of ADC map in MaZda. The abscissa represents different gray values within the ROI, and the ordinate represents the occurrence frequency of each gray value. ADC, apparent diffusion coefficient; ROI, region of interest.
Statistical analyses
Statistical analyses was performed using SPSS v. 11.0 software (SPSS Inc., Chicago, IL). All 11 parameters of the ADC histogram were assessed for normality using the Kolmogorov–Smirnov (K–S) test and test of variance homogeneity. For parameters that conformed to normal distribution and homogeneity of variance, One-way ANOVA was used for comparison among the three groups and an LSD-t test was used for pairwise comparison between groups. The Kruskal–Wallis and Mann–Whitney tests were used to evaluate the other parameters. Receiver operating characteristic (ROC) curve analysis was employed to investigate the diagnostic ability of significant parameters for discrimination. The area under the curve (AUC) was used to evaluate the diagnostic efficiency of histogram parameter. Cut-off values were calculated with the maximum of the Youden index (Youden index = sensitivity + specificity − 1). A probability p value ＜ 0.05 was considered to indicate statistical significance.
Results
The statistical results revealed that only four parameters (ADC_1th, ADC_10th, ADC_90th and skewness) conformed to normal distribution and homogeneity of variance. Thus, for these four parameters, One-way ANOVA and LSD-t test was used for comparison, and Kruskal–Wallis and Mann–Whitney tests were used to evaluate the other parameters.
Table 1 summarizes the detailed comparison of ADC histogram parameters among pleomorphic adenomas, Warthin tumors, and malignant parotid gland tumors. In total, nine parameters (ADC_mean, ADC_minimum, ADC_maximum, ADC_10th, ADC_50th, ADC_90th, ADC_99th, skewness, variance) among the three groups were statistically different (all p < 0.05). No significant differences were found for the other two parameters (ADC_1th and kurtosis). Significant differences were found for all nine parameters between pleomorphic adenomas and Warthin tumors. Significant differences were also found between pleomorphic adenomas and malignant parotid gland tumors (all p < 0.05). However, the comparison of Warthin tumors to malignant parotid gland tumors indicated significant differences for only three parameters (ADC_mean, ADC_50th and skewness) (p < 0.05)
Table 1. .
Differences of ADC histogram parameters among pleomorphic adenomas, warthin tumors and malignant parotid gland tumors
| Parameters | ADC_minimum | ADC_maximum | ADC_mean | Variance | ADC_50th | ADC_99th |
| Pleomorphic adenomas | c67.99±18.27a | 230.95±58.68b | 149.68±29.84b | 804.61±606.68b | 150.92±27.90b | c215.56±23.47b |
| Warthin tumors | 41.50±14.56 | c176.45±29.48 | c110.98±15.26 | c524.61±242.85 | c107.78±14.99 | c178.61±24.95 |
| Malignant parotid gland tumors | c52.67±20.87a | c199.42±47.59a | *126.540±22.23ab | 526.14±592.48a | *124.03±22.19ab | 184.40±28.60a |
| X2 value | 19.060 | 26.676 | 36.110 | 12.187 | 34.997 | 20.528 |
| p-value | 0.000 | 0.000 | 0.000 | 0.002 | 0.000 | 0.000 |
| Parameters | Skewness | ADC_10th | ADC_90th | |||
| Pleomorphic adenomas | c-0.69±0.74b | c123.27±18.20b | c189.36±27.21b | |||
| Warthin tumors | c0.81±0.55 | c88.53±10.51 | c138.39±22.74 | |||
| Malignant parotid gland tumors | c0.14±0.71ab | c109.58±21.60a | c158.53±35.43a | |||
| F value | 25.554 | 29.705 | 17.907 | |||
| p-value | 0.000 | 0.000 | 0.000 |
ADC, apparent diffusion coefficient.; IQR, Interquartile Range; SD, standard deviation.
indicate p < 0.05 when compared with pleomorphic adenomas.
indicate p < 0.05 when compared with Warthin tumors.
indicate the data are consistent with normal distribution and homogeneity of variance are presented as mean ± SD; the other data are presented as median ± IQR.
Tables 2–4 summarize the diagnostic performance of significant ADC histogram parameters in differentiating pleomorphic adenomas from Warthin tumors, pleomorphic adenomas from malignant parotid gland tumors, and Warthin tumors from malignant parotid gland tumors.
Table 2. .
ROC analysis of ADC histogram parameters for discriminating pleomorphic adenomas from Warthin tumors
| Parameters | AUC | Cut-off value | Sensitivity | Specificity | p- value | Youden index |
| ADC_minimum | 0.881 | 50.75 | 86.1% | 78.6% | 0.000 | 0.647 |
| ADC_maximum | 0.933 | 203.42 | 88.9% | 85.7% | 0.000 | 0.746 |
| ADC_mean | 0.976 | 132.00 | 88.9% | 100.0% | 0.000 | 0.889 |
| Variance | 0.786 | 611.33 | 69.4% | 64.3% | 0.002 | 0.337 |
| Skewness | 0.970 | 0.05 | 100.0% | 86.1% | 0.000 | 0.861 |
| ADC_10th | 0.967 | 99.80 | 94.4% | 92.9% | 0.000 | 0.873 |
| ADC_50th | 0.970 | 130.25 | 88.9% | 100.0% | 0.000 | 0.889 |
| ADC_90th | 0.946 | 168.67 | 83.3% | 100.0% | 0.000 | 0.833 |
| ADC_99th | 0.881 | 202.63 | 80.6% | 85.7% | 0.000 | 0.663 |
ADC, apparent diffusion coefficient; AUC, area under the curve; ROC, receiver operating characteristic.
Table 3. .
ROC analysis of ADC histogram parameters for discriminating pleomorphic adenomas from malignant parotid gland tumors
| Parameters | AUC | Cut-off value | Sensitivity | Specificity | p- value | Youden index |
| ADC_minimum | 0.705 | 60.17 | 66.7% | 73.7% | 0.013 | 0.404 |
| ADC_maximum | 0.787 | 209.18 | 86.1% | 73.7% | 0.001 | 0.598 |
| ADC_mean | 0.851 | 132.79 | 88.9% | 73.7% | 0.000 | 0.626 |
| Variance | 0.709 | 679.97 | 61.1% | 68.4% | 0.011 | 0.295 |
| Skewness | 0.781 | −0.29 | 78.9% | 63.9% | 0.001 | 0.428 |
| ADC_10th | 0.866 | 107.04 | 86.1% | 73.7% | 0.000 | 0.598 |
| ADC_50th | 0.841 | 127.42 | 91.7% | 73.7% | 0.000 | 0.654 |
| ADC_90th | 0.787 | 166.87 | 86.1% | 73.7% | 0.001 | 0.598 |
| ADC_99th | 0.760 | 198.50 | 83.3% | 78.9% | 0.002 | 0.622 |
ADC, apparent diffusion coefficient; AUC, area under the curve; ROC, receiver operating characteristic.
Table 4. .
ROC analysis of ADC histogram parameters for distinguishing Warthin tumors from malignant parotid gland tumors
| Parameters | AUC | Cut-off value | Sensitivity | Specificity | p- value | Youden index |
| ADC mean | 0.726 | 122.06 | 63.2% | 78.6% | 0.029 | 0.418 |
| Skewness | 0.782 | 0.20 | 85.7% | 63.2% | 0.006 | 0.489 |
| ADC_50th | 0.737 | 118.50 | 68.4% | 78.6% | 0.022 | 0.470 |
ADC, apparent diffusion coefficient; AUC, area under the curve; ROC, receiver operating characteristic.
In the comparison of pleomorphic adenomas to Warthin tumors, the best diagnostic performance was achieved at the threshold of ADC_mean = 132.00 (AUC = 0.976, sensitivity = 88.9%, specificity = 100.0%, Youden index = 0.889), followed by skewness, ADC_50th and ADC_10th (Figures 4 and 5).
Figure 4. .
ROC curves of ADC_minimum, ADC_maximum, ADC_mean, variance, ADC_10th, ADC_50th, ADC_90th and ADC_99th for differentiating pleomorphic adenomas from Warthin tumors. ADC, apparent diffusion coefficient; ROC, receiver operating characteristic.
Figure 5. .
ROC curves of skewness for differentiating pleomorphic adenomas from Warthin tumors.ADC, apparent diffusion coefficient; ROC, receiver operating characteristic.
In the comparison of pleomorphic adenomas to malignant parotid gland tumors, the best diagnostic performance was achieved at the threshold of ADC_10th = 107.04 (AUC = 0.866, sensitivity = 86.1%, specificity = 73.7%, Youden index = 0.598), followed by ADC_mean, ADC_50th, ADC_90th and ADC_maximum (Figures 6 and 7).
Figure 6. .
ROC curves of ADC_minimum, ADC_maximum, ADC_mean, variance, ADC_10th, ADC_50th, ADC_90th and ADC_99th for differentiating pleomorphic adenomas from malignant parotid gland tumors. ADC, apparent diffusion coefficient; ROC, receiver operating characteristic.
Figure 7. .
ROC curves of skewness for differentiating pleomorphic adenomas from malignant parotid gland tumors. ADC, apparent diffusion coefficient; ROC, receiver operating characteristic.
In the comparison of pleomorphic adenomas to malignant parotid gland tumors, the best diagnostic performance was achieved at the threshold of skewness = 0.20 (AUC = 0.782, sensitivity = 85.7%, specificity = 63.2%, Youden index = 0.489), followed by ADC_mean and ADC_50th (Figures 8 and 9).
Figure 8. .
ROC curves of ADC_mean and ADC_90th, for differentiating Warthin tumors from malignant parotid gland tumors. ADC, apparent diffusion coefficient; ROC, receiver operating characteristic.
Figure 9. .
ROC curves of skewness for differentiating Warthin tumors from malignant parotid gland tumors. ADC, apparent diffusion coefficient; ROC, receiver operating characteristic.
Discussion
The results of our study suggest that nine parameters (ADC_mean, ADC_minimum, ADC_maximum, ADC_10th, ADC_50th, ADC_90th, ADC_99th, skewness, variance) differed significantly among three groups. ADC_mean, ADC_50th and skewness were statistically different between any two groups. The whole tumor histogram statistics (i.e., ADC_mean) obtained from entire tumor volume data appears to have a high value in distinguishing tumor type. In addition to ADC_mean, ADC_50th and skewness were also useful in discriminating pleomorphic adenomas from Warthin tumors. The parameters ADC_10th and ADC_50th were useful for the discriminating pleomorphic adenomas from malignant parotid gland tumors. With the added information of whole-lesion ADC histograms, it may be possible to avoid missing cancers or performing unnecessary surgeries or biopsies for benign lesions in the parotid gland.
DWI has been introduced to distinguish head and neck tumors. Previous researchers have reported the usefulness of combining DWI with a calculation of ADC in the head and neck region in the diagnosis of parotid gland, tongue, and sinonasal lesions.17–21 In our study, the mean ADC of pleomorphic adenomas was much higher than that of malignant tumors, and the mean ADC of Warthin tumors was lower than that of malignant tumors. These results agree with findings of previous reports.6 We explored the utility of multiparametric ADC histogram analysis for differentiating benign from malignant parotid gland tumors. We found that mean ADC derived from the entire tumor shows a significant correlation with tumor type. Our study differed from most previous studies which have been largely based on selected ROIs drawn on the solid part of the tumor for analysis. However, cystic degeneration and tumor necrosis have been reported as common imaging features of parotid gland tumors, both benign or malignant.22
Whole-lesion ADC histogram analysis is a more objective approach which should provide quantitative information about the tissue characteristics and heterogeneity of the whole tumor.23
In our study, ADC_mean, skewness and ADC_50th yielded higher areas under the ROC curve than other parameters for differentiating pleomorphic adenomas from Warthin tumors. By deriving ADC_mean or ADC_50th from selected volume elements, the confounding effect of the high ADCs of necrosis or cystic component may be reduced.23 Skewness represents a measure of asymmetry of the probability distribution and is usually more difficult to interpret.9 Our study has shown that skewness is useful for identifying tumor types; the skewness of malignant tumors is positive and the value is the highest. The significantly higher positive (right) skewness is caused by a small number of pixels with high ADCs and a large number of pixels with low ADCs. This ADC distribution corresponds histologically to large tumor areas with densely packed cells and only minor amounts of scattered necrosis.22 The skewness of pleomorphic adenomas is negative, a finding almost exactly opposite that of malignant tumors. Therefore, measures of skewness may be used as biomarkers of tumor heterogeneity. In addition, we found that the low percentile values (ADC_10th) could be used to differentiate pleomorphic adenomas from malignant tumors. A percentile represents the value below which a percentage of observations is calculated. As suggested by Kang,23 low percentile ADC values measured within a tumor correlate well with areas of high cellularity.
ADC values in this study were obtained by RESOLVE. In prior studies, SS-EPI DWI was the most widely used sequence to measure ADC in head and neck region because of its fast data acquisition. However, image quality of SS-EPI DWIs are not always satisfactory because SS-EPI is susceptible to magnetic susceptibility artifacts.11 Additionally, it has been reported that SS-EPI DWI is vulnerable to transient signal loss and may lead to an overestimation of ADC in parotid glands.24 RESOLVE is a novel technique using a two-dimensional navigator-based reacquisition that performs a nonlinear phase correction and controls the real-time reacquisition of unusable data that cannot be corrected.9,25 As such, RESOLVE has been used to obtain high resolution of DWIs, and with the use of parallel imaging can allow suitable scan time for clinical routine application.26,27 Menglong et al reported that RESOLVE significantly improved the image quality for evaluations of sinonasal lesions and offers more accurate ADC values than SS-EPI.25
The main limitation of this study was the small sample size of malignant tumors. Further, these malignant tumors included a variety of pathological types; we did not undertake statistical analyses among the different pathological types. In addition, morphologic analyses of the lesions were not included. Therefore, future studies of the clinical value of RESOLVE should involve larger sample sizes. The addition of morphologic criteria to RESOLVE may improve diagnostic accuracy.
In conclusion, whole-lesion ADC histogram analysis following high-resolution DWI is effective in differentiating between common parotid gland tumors.
Footnotes
Contributors: Z Z and C S performed data collection analyzed the results. Both of them contributed equally to the writing of this article. J C designed the work and gave the final approval of the version to be published. Y Z assisted with drafting the work and revising it critically for important intellectual content. B W and J X analyzed the results, and was involved in data collection and manuscript preparation.
Contributor Information
Zanxia Zhang, Email: 15838163739@163.com.
Chengru Song, Email: 252936388@163.com.
Yong Zhang, Email: 15838163739@163.com.
Jinxia Zhu, Email: jinxia.zhu@siemens.com.
Jingliang Cheng, Email: chengjl-2008@163.com.
REFERENCES
- 1.Papadogeorgakis N, Skouteris CA, Mylonas AI, Angelopoulos AP. Superficial parotidectomy: technical modifications based on tumour characteristics. J Craniomaxillofac Surg 2004; 32: 350–3. doi: 10.1016/j.jcms.2004.05.004 [DOI] [PubMed] [Google Scholar]
- 2.Zbären P, Nuyens M, Loosli H, Stauffer E. Diagnostic accuracy of fine-needle aspiration cytology and frozen section in primary parotid carcinoma. Cancer 2004; 100: 1876–83. doi: 10.1002/cncr.20186 [DOI] [PubMed] [Google Scholar]
- 3.Seethala RR, LiVolsi VA, Baloch ZW. Relative accuracy of fine-needle aspiration and frozen section in the diagnosis of lesions of the parotid gland. Head Neck 2005; 27: 217–23. doi: 10.1002/hed.20142 [DOI] [PubMed] [Google Scholar]
- 4.Milad P, Elbegiermy M, Shokry T, Mahmoud H, Kamal I, Taha MS, et al. The added value of pretreatment DW MRI in characterization of salivary glands pathologies. Am J Otolaryngol 2017; 38: 13–20. doi: 10.1016/j.amjoto.2016.09.002 [DOI] [PubMed] [Google Scholar]
- 5.Assili S, Fathi Kazerooni A, Aghaghazvini L, Saligheh Rad HR, Pirayesh Islamian J. Dynamic Contrast Magnetic Resonance Imaging (DCE-MRI) and Diffusion Weighted MR Imaging (DWI) for Differentiation between Benign and Malignant Salivary Gland Tumors. J Biomed Phys Eng 2015; 5: 157–68. [PMC free article] [PubMed] [Google Scholar]
- 6.Habermann CR, Arndt C, Graessner J, Diestel L, Petersen KU, Reitmeier F, et al. Diffusion-weighted echo-planar MR imaging of primary parotid gland tumors: is a prediction of different histologic subtypes possible? AJNR Am J Neuroradiol 2009; 30: 591–6. doi: 10.3174/ajnr.A1412 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bull JG, Saunders DE, Clark CA. Discrimination of paediatric brain tumours using apparent diffusion coefficient histograms. Eur Radiol 2012; 22: 447–57. doi: 10.1007/s00330-011-2255-7 [DOI] [PubMed] [Google Scholar]
- 8.Shindo T, Fukukura Y, Umanodan T, Takumi K, Hakamada H, Nakajo M, et al. Histogram Analysis of Apparent Diffusion Coefficient in Differentiating Pancreatic Adenocarcinoma and Neuroendocrine Tumor. Medicine 2016; 95: e2574. doi: 10.1097/MD.0000000000002574 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Just N. Improving tumour heterogeneity MRI assessment with histograms. Br J Cancer 2014; 111: 2205–13. doi: 10.1038/bjc.2014.512 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Porter DA, Heidemann RM. High resolution diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging and a two-dimensional navigator-based reacquisition. Magn Reson Med 2009; 62: 468–75. doi: 10.1002/mrm.22024 [DOI] [PubMed] [Google Scholar]
- 11.Koyasu S, Iima M, Umeoka S, Morisawa N, Porter DA, Ito J, 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: 3088–96. doi: 10.1007/s00330-014-3369-5 [DOI] [PubMed] [Google Scholar]
- 12.Holdsworth SJ, Skare S, Newbould RD, Guzmann R, Blevins NH, Bammer R. Readout-segmented EPI for rapid high resolution diffusion imaging at 3T. Eur J Radiol 2008; 65: 36–46. doi: 10.1016/j.ejrad.2007.09.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bogner W, Pinker K, Zaric O, Baltzer P, Minarikova L, Porter D, et al. Bilateral diffusion-weighted MR imaging of breast tumors with submillimeter resolution using readout-segmented echo-planar imaging at 7 T. Radiology 2015; 274: 74–84. doi: 10.1148/radiol.14132340 [DOI] [PubMed] [Google Scholar]
- 14.Thian YL, Xie W, Porter DA, Weileng Ang B. Readout-segmented echo-planar imaging for diffusion-weighted imaging in the pelvis at 3T-A feasibility study. Acad Radiol 2014; 21: 531–7. doi: 10.1016/j.acra.2014.01.005 [DOI] [PubMed] [Google Scholar]
- 15.Wu J, Wang Q, Zhang J, Wang XN, Liu XS, Zhang YD, 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 2016; 41: 100–8. [DOI] [PubMed] [Google Scholar]
- 16.Fruehwald-Pallamar J, Czerny C, Holzer-Fruehwald L, Nemec SF, Mueller-Mang C, Weber M, et al. Texture-based and diffusion-weighted discrimination of parotid gland lesions on MR images at 3.0 Tesla. NMR Biomed 2013; 26: 1372–9. doi: 10.1002/nbm.2962 [DOI] [PubMed] [Google Scholar]
- 17.Minosse S, Marzi S, Piludu F, Vidiri A. Correlation study between DKI and conventional DWI in brain and head and neck tumors. Magn Reson Imaging 2017; 42: 114–22. doi: 10.1016/j.mri.2017.06.006 [DOI] [PubMed] [Google Scholar]
- 18.Ailianou A, Mundada P, De Perrot T, Pusztaszieri M, Poletti P-A, Becker M. MRI with DWI for the Detection of Posttreatment Head and Neck Squamous Cell Carcinoma: Why Morphologic MRI Criteria Matter. AJNR Am J Neuroradiol 2018; 39: 748–55. doi: 10.3174/ajnr.A5548 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Li S, Cheng J, Zhang Y, Zhang Z. Differentiation of benign and malignant lesions of the tongue by using diffusion-weighted MRI at 3.0 T. Dentomaxillofac Radiol 2015; 44: 20140325. doi: 10.1259/dmfr.20140325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Driessen JP, van Kempen PMW, van der Heijden GJ, Philippens MEP, Pameijer FA, Stegeman I, et al. Diffusion-weighted imaging in head and neck squamous cell carcinomas: a systematic review. Head Neck 2015; 37: 440–8. doi: 10.1002/hed.23575 [DOI] [PubMed] [Google Scholar]
- 21.Xiao Z, Zhong Y, Tang Z, Qiang J, Qian W, Wang R, et al. Standard diffusion-weighted, diffusion kurtosis and intravoxel incoherent motion MR imaging of sinonasal malignancies: correlations with Ki-67 proliferation status. Eur Radiol 2018; 28: 2923–33. doi: 10.1007/s00330-017-5286-x [DOI] [PubMed] [Google Scholar]
- 22.Kato H, Kanematsu M, Watanabe H, Mizuta K, Aoki M. Salivary gland tumors of the parotid gland: CT and MR imaging findings with emphasis on intratumoral cystic components. Neuroradiology 2014; 56: 789–95. doi: 10.1007/s00234-014-1386-3 [DOI] [PubMed] [Google Scholar]
- 23.Kang Y, Choi SH, Kim Y-J, Kim KG, Sohn C-H, Kim J-H, et al. Gliomas: Histogram analysis of apparent diffusion coefficient maps with standard- or high-b-value diffusion-weighted MR imaging--correlation with tumor grade. Radiology 2011; 261: 882–90. doi: 10.1148/radiol.11110686 [DOI] [PubMed] [Google Scholar]
- 24.Liu Y-J, Lee Y-H, Chang H-C, Huang T-Y, Chiu H-C, Wang C-W, et al. A potential risk of overestimating apparent diffusion coefficient in parotid glands. PLoS One 2015; 10: e0124118. doi: 10.1371/journal.pone.0124118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zhao M, Liu Z, Sha Y, Wang S, Ye X, Pan Y, 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: 166–72. doi: 10.1016/j.mri.2015.10.010 [DOI] [PubMed] [Google Scholar]
- 26.Holdsworth SJ, Yeom K, Skare S, Gentles AJ, Barnes PD, Bammer R, et al. Clinical application of readout-segmented- echo-planar imaging for diffusion-weighted imaging in pediatric brain. AJNR Am J Neuroradiol 2011; 32: 1274–9. doi: 10.3174/ajnr.A2481 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bogner W, Pinker-Domenig K, Bickel H, Chmelik M, Weber M, Helbich TH, 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. doi: 10.1148/radiol.12111494 [DOI] [PubMed] [Google Scholar]