Fully automated 3D machine learning model for HPV status characterization in oropharyngeal squamous cell carcinomas based on CT images.

BACKGROUND: Human papillomavirus (HPV) status plays a major role in predicting oropharyngeal squamous cell carcinoma (OPSCC) survival. This study assesses the accuracy of a fully automated 3D convolutional neural network (CNN) in predicting HPV status using CT images. METHODS: Pretreatment CT images from OPSCC patients were used to train a 3D DenseNet-121 model to predict HPV-p16 status. Performance was evaluated by the ROC Curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score. RESULTS: The network achieved a mean AUC of 0.80 ± 0.06. The best-preforming fold had a sensitivity of 0.86 and specificity of 0.92 at the Youden's index. The PPV, NPV, and F1 scores are 0.97, 0.71, and 0.82, respectively. CONCLUSIONS: A fully automated CNN can characterize the HPV status of OPSCC patients with high sensitivity and specificity. Further refinement of this algorithm has the potential to provide a non-invasive tool to guide clinical management.

Differentiation of salivary gland tumours using diffusion-weighted image-based virtual MR elastography: a pilot study.

OBJECTIVES: Differentiation among benign salivary gland tumours, Warthin tumours (WTs), and malignant salivary gland tumours is crucial to treatment planning and predicting patient prognosis. However, differentiation of those tumours using imaging findings remains difficult. This study evaluated the usefulness of elasticity determined from diffusion-weighted image (DWI)-based virtual MR elastography (MRE) compared with conventional magnetic resonance imaging (MRI) findings in differentiating the tumours. METHODS: This study included 17 benign salivary gland tumours, 6 WTs, and 11 malignant salivary gland tumours scanned on neck MRI. The long and short diameters, T1 and T2 signal intensities, tumour margins, apparent diffusion coefficient (ADC) values, and elasticity from DWI-based virtual MRE of the tumours were evaluated. The interobserver agreement in measuring tumour elasticity and the receiver operating characteristic (ROC) curves were also assessed. RESULTS: The long and short diameters and the T1 and T2 signal intensities showed no significant difference among the 3 tumour groups. Tumour margins and the mean ADC values showed significant differences among some tumour groups. The elasticity from virtual MRE showed significant differences among all 3 tumour groups and the interobserver agreement was excellent. The area under the ROC curves of the elasticity were higher than those of tumour margins and mean ADC values. CONCLUSION: Elasticity values based on DWI-based virtual MRE of benign salivary gland tumours, WTs, and malignant salivary gland tumours were significantly different. The elasticity of WTs was the highest and that of benign tumours was the lowest. The elasticity from DWI-based virtual MRE may aid in the differential diagnosis of salivary gland tumours.

Bright diffusion sign: A sensitive and specific radiologic biomarker for multinodular and vacuolating neuronal tumor.

BACKGROUND AND PURPOSE: Accurate differentiation between multinodular and vacuolating neuronal tumor (MVNT) and dysembryoplastic neuroepithelial tumor (DNET) is important for treatment decision-making. We aimed to develop an accurate radiologic diagnostic model for differentiating MVNT from DNET using T2WI and diffusion-weighted imaging (DWI). MATERIALS AND METHODS: A total of 56 patients (mean age, 47.48±17.78 years; 31 women) diagnosed with MVNT (n = 37) or DNET (n = 19) who underwent brain MRI, including T2WI and DWI, were included. Two board-certified neuroradiologists performed qualitative (bubble appearance, cortical involvement, bright diffusion sign, and bright apparent diffusion coefficient [ADC] sign) and quantitative (nDWI and nADC) assessments. A diagnostic tree model was developed with significant and reliable imaging findings using an exhaustive chi-squared Automatic Interaction Detector (CHAID) algorithm. RESULTS: In visual assessment, the imaging features that showed high diagnostic accuracy and interobserver reliability were the bright diffusion sign and absence of cortical involvement (bright diffusion sign: accuracy, 94.64 %; sensitivity, 91.89 %; specificity, 100.00 %; interobserver agreement, 1.00; absence of cortical involvement: accuracy, 92.86 %; sensitivity, 89.19 %; specificity, 100.00 %; interobserver agreement, 1.00). In quantitative analysis, nDWI was significantly higher in MVNT than in DENT (1.52 ± 0.34 vs. 0.91 ± 0.27, p < 0.001), but the interobserver agreement was fair (intraclass correlation coefficient = 0.321). The overall diagnostic accuracy of the tree model with visual assessment parameters was 98.21 % (55/56). CONCLUSION: The bright diffusion sign and absence of cortical involvement are accurate and reliable imaging findings for differentiating MVNT from DNET. By using simple, intuitive, and reliable imaging findings, such as the bright diffusion sign, MVNT can be accurately differentiated from DNET.

MR Findings of Fungus Ball: Significance of High Signal Intensity on T1-Weighted Images.

BACKGROUND: Central dark-signal intensity with high-signal, hypertrophic mucosal wall of paranasal sinuses on T2-weighted images (T2WI) is a characteristic magnetic resonance imaging (MRI) feature of sinonasal fungus ball. However, this finding is usually interpreted as non-fungal chronic sinusitis with central normal sinus air. In addition, T1-weighted images (T1WI) and T2WI are basic sequences of all magnetic resonance (MR) examinations. Therefore, we evaluated the usefulness of T1WI for detecting fungus balls comparing with computed tomography (CT) findings and T2-weighted MRI findings. METHODS: This retrospective study was approved by the Institutional Review Board of Korea University Guro Hospital. Two reviewers assessed preoperative CT and MR images of 55 patients with pathologically confirmed fungus balls. Reviewers evaluated the presence and patterns of calcifications on CT. Overall signals and the presence and extent of certain signals of fungus balls on MRI were also assessed. The relationship between calcifications and MRI signals was also evaluated. RESULTS: Of the patients, 89.1% had calcifications on CT. All had dark signal portions with high signal, hypertrophic mucosal walls on T2WI. Most (92.7%) patients showed iso- to hyper-intense overall signals on T1WI and 89.1% had T1-weighted high signal portions on MRI. The presence, patterns, and location of calcifications had no significant correlation with T1-weighted high-signal intensity portion. CONCLUSION: Fungus ball can be suggested by the presence of the hyper-signal intensity portions in the fungal mass on T1WI in conjunction with dark-signal lesions surrounded by high-signal, hypertrophic mucosal walls in paranasal sinuses on T2WI.

MRI Findings of Spontaneous Intracranial Hypotension: Usefulness of Straight Sinus Distention.

OBJECTIVE. Spontaneous intracranial hypotension (SIH) shows various characteristic MRI findings. We evaluated the usefulness of straight sinus distention compared with transverse sinus distention and also evaluated other MRI findings of SIH. MATERIALS AND METHODS. Forty-three consecutive patients (28 female and 15 male patients) treated for SIH and 43 age- and sex-matched control subjects at two institutions from 2012 through 2014 were included in this study. Two reviewers determined whether the transverse sinus distention sign and straight sinus distention sign were present on MRI. Diagnostic performance values and interobserver agreement were calculated. Reviewers also assessed MRI examinations in consensus for the presence of the following findings: pachymeningeal enhancement, subdural effusion or hematoma, enlargement of the pituitary gland, and downward displacement of the brainstem and tonsils. RESULTS. The sensitivity, specificity, and diagnostic accuracy of the transverse sinus distention sign for SIH were 76.7%, 83.7%, and 80.2%, whereas those of the straight sinus distention sign were 79.1%, 95.4%, and 87.2%, respectively. The specificity of the straight sinus distention sign for SIH was significantly higher (p = 0.025) than that of the transverse sinus distention sign. In addition, the straight sinus distention sign showed substantial agreement (κ = 0.79), whereas the transverse sinus distention sign showed moderate agreement (κ = 0.60). The diagnostic accuracy of the presence of either transverse or straight sinus distention (83.7%) was significantly higher than that of pachymeningeal enhancement (80.2%, p = 0.032). CONCLUSION. The straight sinus distention sign could be helpful for the diagnosis of SIH because it has sensitivity comparable to other imaging findings and higher specificity and higher level of interobserver agreement than other imaging findings.

Analysis of postoperative ultrasonography surveillance after total thyroidectomy in patients with papillary thyroid carcinoma: a multicenter study.

Background In papillary thyroid carcinoma (PTC) patients, postoperative ultrasonography (US) surveillance is recommended at intervals of six or 12 months even though evidence is weak. Purpose To determine the optimal interval of postoperative US surveillance in patients after total thyroidectomy for the treatment of PTC using multicenter data. Material and Methods A total of 1400 patients from seven tertiary hospitals who underwent total thyroidectomy for treatment of PTC were included. Based on a retrospective review of clinical and pathologic results, multiple factors were analyzed according to recurrence/persistence or no recurrence/persistence. In recurrence/persistence group, the interval and number of follow-up US sessions in the initial detection of recurrence/persistence were investigated. Results Of the 1400 patients, 115 (8.2%) showed tumor recurrence/persistence on follow-up US. Of 115 recurrence/persistence cases, 89 (77.4%) were initially detected on US: nodal recurrence/persistence (n = 92), non-nodal recurrence/persistence (n = 22), and both (n = 1). Among the clinical and pathologic factors, only tumor size and N stage were significant predictors for recurrence/persistence. In the recurrence/persistence group, the mean interval and number of follow-up US sessions in the initial detection of recurrence/persistence was 22.3 ± 16.8 months and 2.2 ± 1.9, respectively. Approximately two-thirds of recurrence/persistence cases (76/115, 66.1%) were detected in follow-up US within two years after total thyroidectomy. Conclusion In PTC patients after total thyroidectomy, the optimal interval of the first US follow-up may be one to two years after thyroid surgery, and the appropriate number of postoperative US surveillance sessions within the first five years may be only one or two.

Vascular Pattern Analysis on Microvascular Sonography for Differentiation of Pleomorphic Adenomas and Warthin Tumors of Salivary Glands.

OBJECTIVES: Pleomorphic adenomas and Warthin tumors are the most common salivary gland tumors. It is important to differentiate between them because at least a partial parotidectomy is necessary for pleomorphic adenomas, whereas enucleation is sufficient for Warthin tumors. This study aimed to evaluate the usefulness of vascular pattern analysis using microvascular sonography to differentiate between the tumors. METHODS: Sixty-two patients with pathologically proven pleomorphic adenomas (n = 38) and Warthin tumors (n = 24) were included. For all tumors, grayscale, power Doppler, and microvascular sonographic examinations were performed. Differences in vascular patterns (vascular distribution and internal vascularity) on power Doppler and microvascular sonography as well as grayscale sonographic features (size, shape, border, echogenicity, heterogeneity, and cystic change) between pleomorphic adenomas and Warthin tumors were evaluated. A comparison of diagnostic performances of grayscale sonography with power Doppler sonography and grayscale sonography with microvascular sonography was performed. The level of interobserver agreement between 2 reviewers in diagnosing tumors was evaluated. RESULTS: No grayscale sonographic features showed a significant difference between the tumors. Vascular distributions and internal vascularity on power Doppler sonography (P = .01 and .002) and microvascular sonography (both P < .001) were all significantly different. The diagnostic accuracy of grayscale sonography with microvascular sonography (79.0%) was higher than that of grayscale sonography with power Doppler sonography (72.6%). This difference was significant according to the McNemar test (P = .004). Interobserver agreement was excellent in diagnosing tumors on both grayscale sonography with power Doppler sonography (κ = 0.83) and grayscale sonography with microvascular sonography (κ = 0.94). CONCLUSIONS: Vascular pattern analysis using microvascular sonography with other sonographic features is helpful for differentiating between pleomorphic adenomas and Warthin tumors.

A semi-automated volumetric software for segmentation and perfusion parameter quantification of brain tumors using 320-row multidetector computed tomography: a validation study.

PURPOSE: We developed a semi-automated volumetric software, NPerfusion, to segment brain tumors and quantify perfusion parameters on whole-brain CT perfusion (WBCTP) images. The purpose of this study was to assess the feasibility of the software and to validate its performance compared with manual segmentation. METHODS: Twenty-nine patients with pathologically proven brain tumors who underwent preoperative WBCTP between August 2012 and February 2015 were included. Three perfusion parameters, arterial flow (AF), equivalent blood volume (EBV), and Patlak flow (PF, which is a measure of permeability of capillaries), of brain tumors were generated by a commercial software and then quantified volumetrically by NPerfusion, which also semi-automatically segmented tumor boundaries. The quantification was validated by comparison with that of manual segmentation in terms of the concordance correlation coefficient and Bland-Altman analysis. RESULTS: With NPerfusion, we successfully performed segmentation and quantified whole volumetric perfusion parameters of all 29 brain tumors that showed consistent perfusion trends with previous studies. The validation of the perfusion parameter quantification exhibited almost perfect agreement with manual segmentation, with Lin concordance correlation coefficients (ρ c) for AF, EBV, and PF of 0.9988, 0.9994, and 0.9976, respectively. On Bland-Altman analysis, most differences between this software and manual segmentation on the commercial software were within the limit of agreement. CONCLUSIONS: NPerfusion successfully performs segmentation of brain tumors and calculates perfusion parameters of brain tumors. We validated this semi-automated segmentation software by comparing it with manual segmentation. NPerfusion can be used to calculate volumetric perfusion parameters of brain tumors from WBCTP.

Postoperative ultrasonography surveillance in patients with follicular thyroid carcinoma: a multicenter study.

OBJECTIVE: This study aimed to evaluate the locoregional recurrence rate of follicular thyroid carcinoma (FTC) and to assess the appropriate frequency of postoperative ultrasonography (US) surveillance for detecting tumor recurrence. METHODS: The review boards of the seven participating institutions approved this study. From 2000 to 2011, 186 patients underwent at postoperative US at least once; US was performed by experienced radiologists at each institution. Based on the US and histopathological results, locoregional tumor recurrence was assessed. RESULTS: The T stages of the 186 patients were T1a (8.1%), T1b (21.5%), T2 (39.8%), T3 (30.6%), T4a (0%), and T4b (0%). The N stages were unknown (24.2%), N0 (71.5%), N1a (3.2%), and N1b (1.1%), and the M stages unknown (29.6%), M0 (66.1%), and M1 (4.3%). Tumors recurred in only 6 (3.2%) patients during the follow-up period over 5 years. Among them, no patients showed the initial suspicion of recurrences on routine follow-up US. The session number and interval of postoperative US differed significantly between patients with recurrence and those without recurrence. The mean interval of postoperative follow-up US at the first detection time of tumor recurrence was 37.5 ± 18.5 months (range 9-62 months). Significantly more FTCs were at an advanced N and M stage in the recurrence group than in the non-recurrence group (p < 0.05). CONCLUSIONS: Routine postoperative US surveillance may be unnecessary for detecting tumor recurrence after thyroid surgery in FTC patients.

Off-site evaluation of three-dimensional ultrasound for the diagnosis of thyroid nodules: comparison with two-dimensional ultrasound.

OBJECTIVES: We compared the diagnostic performance of off-site evaluation between prospectively obtained 3D and 2D ultrasound for thyroid nodules. METHODS: 3D and 2D ultrasonographies were preoperatively obtained from 85 consecutive patients (mean age, 51 years; age range, 28-83 years) who were referred for a total thyroidectomy. Three radiologists independently evaluated 3D and 2D images of 91 pathologically confirmed thyroid nodules (30 benign and 61 malignant nodules) for nodule characterization. Diagnostic performance, interobserver agreement and time for scanning were compared between 3D and 2D. RESULTS: 3D had significantly higher sensitivities than 2D for predicting malignancy (78.7 % vs. 61.2 %, P < 0.01) and extrathyroidal extension (66.7 % vs. 46.4 %, P = 0.03) in malignancy. In terms of specificities, there were no statistically significant differences between 2D and 3D for predicting malignancy (78.4 % vs. 74.8 %, P = 1.00) and extrathyroidal extension (63.6 % vs. 57.6 %, P = 0.46). With respect to interobserver agreement, 3D showed moderate agreement (κ = 0.53) for predicting extrathyroidal extension in malignancy compared with 2D ultrasound, which showed fair agreement (κ = 0.37). 3D saved time (30 ± 56.52 s) for scanning compared with 2D. CONCLUSION: For off-site evaluation, 3D US is more useful for diagnosis of thyroid nodules than 2D US. KEY POINTS: • 3D had higher sensitivity than 2D for predicting malignancy and extrathyroidal extension. • 3D showed better agreement for predicting extrathyroidal extension in malignancy than 2D. • 3D thyroid ultrasound saved time for scanning compared with 2D. • For off-site evaluation of thyroid nodules, 3D is more useful than 2D.

Evaluation of the microenvironmental heterogeneity in high-grade gliomas with IDH1/2 gene mutation using histogram analysis of diffusion-weighted imaging and dynamic-susceptibility contrast perfusion imaging.

The purpose of our study was to explore the difference between isocitrate dehydrogenase (IDH)-1/2 gene mutation-positive and -negative high-grade gliomas (HGGs) using histogram analysis of apparent diffusion coefficient (ADC) and normalized cerebral blood volume (nCBV) maps. We enrolled 52 patients with histopathologically confirmed HGGs with IDH1/2 (P) (n = 16) or IDH1/2 (N) (n = 36). Histogram parameters of ADC and nCBV maps were correlated with gene mutations by using the unpaired student's t test and multivariable stepwise logistic regression analysis. The mean ADC value was higher in the IDH1 (P) group than IDH1 (N) (1,282.8 vs. 1,159.6 mm(2)/s, P = .0113). In terms of the cumulative ADC histograms, the 10th and 50th percentile values were also higher in the IDH1 (P) than IDH1 (N) (P = .0104 and .0183, respectively). We observed a higher 90th percentile value (3.121 vs. 2.397, P = .0208) and a steeper slope between the 10th (C10) and 90th (C90) of cumulative nCBV histograms (0.03386 vs. 0.02425/%, P = .0067) in the IDH1 (N) group. Multivariate analysis showed that the mean ADC mean value (P = .0048), the C90 value (P = .0113), and the slope between C10 and C90 (P = .0049) were the significant variables in the differentiation of IDH1 (P) from IDH1 (N). In conclusion, histogram analysis of ADC and nCBV maps based on entire tumor volume can be a useful tool for distinguishing IDH1 (P) and IDH1 (N), and it predicts that IDH (P) tumors have a more heterogeneous microenvironment than IDH (N) ones.

Differentiation of true progression from pseudoprogression in glioblastoma treated with radiation therapy and concomitant temozolomide: comparison study of standard and high-b-value diffusion-weighted imaging.

PURPOSE: To explore the role of histogram analysis of apparent diffusion coefficient (ADC) maps obtained at standard- and high-b-value (1000 and 3000 sec/mm(2), respectively) diffusion-weighted (DW) imaging in the differentiation of true progression from pseudoprogression in glioblastoma treated with radiation therapy and concomitant temozolomide. MATERIALS AND METHODS: This retrospective study was approved by the institutional review board of Seoul National University Hospital, and informed consent requirement was waived. Thirty patients with histopathologically proved glioblastoma who had undergone concurrent chemotherapy and radiation therapy (CCRT) with temozolomide underwent diffusion-weighted MR imaging with b values of 1000 and 3000 sec/mm(2), and corresponding ADC maps were calculated from entire newly developed or enlarged enhancing lesions after completion of CCRT. Histogram parameters of each ADC map between true progression (n = 15) and pseudoprogression (n = 15) groups were compared by using the unpaired Student t test. Receiver operating characteristic analysis was used to determine the best cutoff values for predictors in the differentiation of true progression from pseudoprogression. Results were validated in an independent test set of nine patients by using the best cutoff value to predict differentiation of true progression from pseudoprogression. The accuracy of the selected best cutoff value in the independent test set was then calculated. RESULTS: In terms of cumulative histograms, the fifth percentile of both ADC at b value of 1000 sec/mm(2) (ADC1000) and the ADC at b value of 3000 sec/mm(2) (ADC3000) were significantly lower in the true progression group than in the pseudoprogression group (P = .049 and P < .001, respectively). In contrast, neither the mean ADC1000 nor the mean ADC3000 was significantly different between the two groups. The diagnostic values of the parameters derived from ADC1000 and ADC3000 were compared, and a significant difference (0.224, P = .016) was found between the area under the receiver operating characteristic curve of the fifth percentile for ADC1000 and that for ADC3000. The accuracies were 66.7% (six of nine patients) and 88.9% (eight of nine patients) based on the fifth percentile of both ADC1000 and ADC3000 in the independent test set, respectively. CONCLUSION: The fifth percentile of the cumulative ADC histogram obtained at a high b value was the most promising parameter in the differentiation of true progression from pseudoprogression of the newly developed or enlarged enhancing lesions after CCRT with temozolomide for glioblastoma treatment. Online supplemental material is available for this article.

Application of symmetry evaluation to deep learning algorithm in detection of mastoiditis on mastoid radiographs.

As many human organs exist in pairs or have symmetric appearance and loss of symmetry may indicate pathology, symmetry evaluation on medical images is very important and has been routinely performed in diagnosis of diseases and pretreatment evaluation. Therefore, applying symmetry evaluation function to deep learning algorithms in interpreting medical images is essential, especially for the organs that have significant inter-individual variation but bilateral symmetry in a person, such as mastoid air cells. In this study, we developed a deep learning algorithm to detect bilateral mastoid abnormalities simultaneously on mastoid anterior-posterior (AP) views with symmetry evaluation. The developed algorithm showed better diagnostic performance in diagnosing mastoiditis on mastoid AP views than the algorithm trained by single-side mastoid radiographs without symmetry evaluation and similar to superior diagnostic performance to head and neck radiologists. The results of this study show the possibility of evaluating symmetry in medical images with deep learning algorithms.

Lesion detectability on diffusion-weighted imaging in transient global amnesia: the influence of imaging timing and magnetic field strength.

INTRODUCTION: Transient global amnesia (TGA) is a benign disease with a sudden-onset, transient memory disturbance. Characteristic punctate high-signal intensity lesions in the hippocampus on diffusion-weighted imaging (DWI) have been reported in variable frequencies. We investigated the influence of the timing of DWI and the magnetic field strength on the lesion detectability in TGA. METHODS: Seventy-three patients diagnosed with TGA underwent DWI within 24 h after the symptom onset and again on day 3 at either 1.5 (n = 31) or 3 T (n = 42). The patients were divided into three subgroups according to the time lapse after the symptom onset to the first DWI (0-6 h, 6-12 h, and 12-24 h). RESULTS: The detection rate of the lesions during the first 24 h rose statistically significantly with increased time lapse after the symptom onset (34% in 0-6 h, 62% in 6-12 h, and 67% in 12-24 h). It increased up to 75% on day 3. The detection rate was higher at 3 T than at 1.5 T in all time points (41% vs. 27% in 0-6 h, 70% vs. 44% in 6-12 h, 80% vs. 57% in 12-24 h, and 86% vs. 61% on day 3), but the statistical significance was achieved only on day 3. CONCLUSION: Awareness of the different lesion detectability on DWI according to the time lapse after the symptom onset can help in diagnosing the patients with suspected TGA. High field strength is another important factor to increase the lesion detectability on DWI.

Imaging finding and analysis of brain lymphoma in contrast-enhanced fluid attenuated inversion recovery sequence.

OBJECTIVE: The purpose of this retrospective study was to report and analyze the image findings of contrast-enhanced fluid-attenuated inversion recovery (CE-FLAIR) sequence of lymphoma in the brain. MATERIAL AND METHODS: Thirty-two immunocompetent patients with biopsy-proven diffuse large B-cell type lymphoma in the brain were evaluated with pre-treatment MRI examinations from August 2014 to April 2020. As stereotactic studies on the day of biopsy, FLAIR and T1-weighted axial images were acquired in 2 mm thickness, before and after administrating gadolinium-based contrast agents, with 3.0 Tesla MR machines. Respective subtraction images were also obtained for both CE-FLAIR and contrast-enhanced T1-wieghted image (CE-T1WI) sequences. The imaging findings, especially the enhancement pattern on CE-FLAIR sequence, were analyzed qualitatively and quantitatively, using semi-automatic segmentation. RESULTS: On CE-FLAIR images, brain lymphomas were poorly enhanced, while showing peripheral rim enhancement (54 of 58 lesions, 93.1 %) and central enhancing foci (40 of 58 lesions, 69.0 %). Seventy percent of central enhancing foci were correlated to areas with low signal intensity on CE-T1WI. In quantitative analysis, the mean signal intensity of CE-T1WI subtraction was 490.44 and that of FLAIR subtraction was 206.13. The standard deviation of all signal intensity values in CE-T1WI subtraction sequence was 143.45, while that of CE-FLAIR subtraction sequence was 118.41. CONCLUSION: On CE-FLAIR, brain lymphomas showed relatively poor and homogeneous enhancement, when compared to CE-T1WI. Most brain lymphomas displayed peripheral rim enhancement and central enhancing foci.

Evaluating the Elasticity of Metastatic Cervical Lymph Nodes in Head and Neck Squamous Cell Carcinoma Patients Using DWI-based Virtual MR Elastography.

PURPOSE: The assessment of metastatic cervical lymph nodes in head and neck squamous cell carcinoma patients is crucial; as such, many studies focusing on non-invasive imaging techniques to evaluate metastatic cervical lymph nodes have been performed. The aim of our study was to assess the usefulness of elasticity values on diffusion weighted imaging (DWI)-based virtual MR elastography in the evaluation of metastatic cervical lymph nodes from head and neck squamous cell carcinoma. METHODS: Two head and neck radiologists measured the elasticity values of 16 metastatic cervical lymph nodes from head and neck squamous cell carcinoma and 13 benign cervical lymph nodes on DWI-based virtual MR elastography maps. Mean, minimum, maximum, and median elasticity values were evaluated for lymph nodes between the two groups and interobserver agreement in measuring the elasticity was also evaluated. RESULTS: The mean, maximum, and median elasticity values of metastatic cervical lymph nodes were significantly higher than those of benign cervical lymph nodes (P = 0.001, 0.01, and 0.002, respectively). Diagnostic accuracy, sensitivity, and specificity of the mean elasticity were 82.8%, 93.8%, and 69.2%, respectively. Interobserver agreement was excellent for the mean and median elasticity (intraclass correlation coefficients were 0.98 for both). CONCLUSION: Estimated elasticity values based on DWI-based virtual MR elastography show significant difference between benign and metastatic cervical lymph nodes from head and neck squamous cell carcinoma. While precise modulation of MR sequences and calibration parameters still needs to be established, elasticity values can be useful in differentiating between these lymph nodes.

Usefulness of 3D High-resolution Vessel Wall MRI in Diffuse Nonaneurysmal SAH Patients.

PURPOSE: In 15-20% of patients with nontraumatic diffuse subarachnoid hemorrhage (SAH), the initial conventional angiography does not reveal a causative vascular abnormality, such as intracranial aneurysm. In this study, we evaluated clinical utility of 3D high-resolution vessel wall magnetic resonance imaging (HR-VWI) in patients with diffuse nonaneurysmal SAH. METHODS: A total of 17 patients with diffuse nonaneurysmal SAH were included in this retrospective study. We characterized demographics and HR-VWI findings and reviewed the clinical management and outcomes. RESULTS: Of the patients 14 (14/17; 82.4%,) showed abnormal findings on HR-VWI, including 5 with intracranial dissections (29.4%), 3 with blood blister-like aneurysm (17.6%), 1 with ruptured fusiform aneurysm (5.9%), and 5 with focal nodular wall enhancement without unclassified pathology (29.4%). Of these patients were treated with endovascular management. Most patients (16/17) had a favorable modified Rankin scale scores of 0-2 on discharge. CONCLUSION: The 3D HR-VWI revealed various hidden pathologies, such as intracranial arterial dissection, blood blister-like aneurysm, and fusiform aneurysm in patients with diffuse nonaneurysmal SAH. In addition, 3D HR-VWI had an impact on the management of SAH. The 3D HR-VWI can be a complementary diagnostic method for patients with diffuse nonaneurysmal SAH in a research or clinical setting.

Rathke Cleft Cysts with Apoplexy-Like Symptoms: Clinicoradiologic Comparisons with Pituitary Adenomas with Apoplexy.

BACKGROUND: Rathke cleft cyst (RCC) can cause acute symptoms mimicking pituitary adenoma (PA) apoplexy. We evaluated the clinicoradiologic features for distinguishing RCC from PA apoplexy. METHODS: We retrospectively evaluated 22 patients with RCC and 24 patients with PA with apoplexy-like symptoms who underwent surgery via a transsphenoidal approach between November 1999 and December 2016. We compared the clinical data and magnetic resonance (MR) images between the 2 groups. RESULTS: The RCC group was younger and had smaller tumors compared with the PA group (P = 0.02 and 0.001, respectively). The incidences of visual deficits and cranial nerve palsy were lower in the RCCs than in the PAs (P ≤ 0.02 for all). MR images showed more frequent intracystic nodules in the RCCs (P < 0.001), whereas nodular enhancement and lateral deviation of the pituitary stalk were more commonly seen in the PAs (P ≤ 0.003 for both). However, the presence of endocrine dysfunction or decreased consciousness, and the recurrence ratio, were not significantly different between the groups (P ≥ 0.48 for all). In the multivariable logistic regression analysis, patients without nodular enhancement had a 15.84-fold greater risk of RCC than did those with nodular enhancement (P = 0.031). The probability of RCC decreased 0.59-fold with each 1-cm3 increase in tumor volume. CONCLUSIONS: RCC with apoplexy-like symptoms has different clinicoradiologic features compared with PA apoplexy. Patients with RCC present with milder ocular symptoms and smaller tumor volumes compared with those with PA apoplexy. The absence of nodular enhancement on MR images could suggest RCC.

Performance of deep learning to detect mastoiditis using multiple conventional radiographs of mastoid.

OBJECTIVES: This study aimed to compare the diagnostic performance of deep learning algorithm trained by single view (anterior-posterior (AP) or lateral view) with that trained by multiple views (both views together) in diagnosis of mastoiditis on mastoid series and compare the diagnostic performance between the algorithm and radiologists. METHODS: Total 9,988 mastoid series (AP and lateral views) were classified as normal or abnormal (mastoiditis) based on radiographic findings. Among them 792 image sets with temporal bone CT were classified as the gold standard test set and remaining sets were randomly divided into training (n = 8,276) and validation (n = 920) sets by 9:1 for developing a deep learning algorithm. Temporal (n = 294) and geographic (n = 308) external test sets were also collected. Diagnostic performance of deep learning algorithm trained by single view was compared with that trained by multiple views. Diagnostic performance of the algorithm and two radiologists was assessed. Inter-observer agreement between the algorithm and radiologists and between two radiologists was calculated. RESULTS: Area under the receiver operating characteristic curves of algorithm using multiple views (0.971, 0.978, and 0.965 for gold standard, temporal, and geographic external test sets, respectively) showed higher values than those using single view (0.964/0.953, 0.952/0.961, and 0.961/0.942 for AP view/lateral view of gold standard, temporal external, and geographic external test sets, respectively) in all test sets. The algorithm showed statistically significant higher specificity compared with radiologists (p = 0.018 and 0.012). There was substantial agreement between the algorithm and two radiologists and between two radiologists (κ = 0.79, 0.8, and 0.76). CONCLUSION: The deep learning algorithm trained by multiple views showed better performance than that trained by single view. The diagnostic performance of the algorithm for detecting mastoiditis on mastoid series was similar to or higher than that of radiologists.

Diagnostic Accuracy and Confidence of [18F] FDG PET/MRI in comparison with PET or MRI alone in Head and Neck Cancer.

The usefulness of PET/MRI in head and neck malignancy has not been fully elucidated. The purpose of our study was to evaluate the diagnostic accuracy and confidence of PET/MRI in comparison with PET or MRI alone. This study included 73 consecutive patients who underwent [18F] FDG PET/MRI in head and neck under the suspicion of malignancy. A neuroradiologist and a nuclear medicine specialist reviewed MRI and PET images, respectively and independently, followed by a consensus review of PET/MRI one month later. For 134 lesions, accuracy and confidence were compared among PET, MRI, and PET/MRI. For lesion base, PET/MRI had a sensitivity of 85.7%, a specificity of 89.1%, a PPV of 89.6%, a negative predictive value of 85.1%, and an accuracy of 87.3%. AUCs of PET/MRI per lesion (0.926) and per patient (0.934) for diagnosing malignancy were higher than PET (0.847 and 0.747, respectively) or MRI (0.836 and 0.798, respectively) alone (P < 0.05). More than 80% of the cases (111/134) showed diagnostic concordance between PET and MRI. PPV of PET/MRI was higher in malignant concordant cases (93.2%, 55/59) than in discordant cases (62.5%, 5/8) (p = 0.040). Confident scoring rate in malignant concordant cases was higher on PET/MRI (96.6%, 57/59) than on MRI (76.3%, 45/59) (p = 0.003). In conclusion, compared with PET or MRI alone, PET/MRI presents better diagnostic performance in accuracy and confidence for diagnosis of malignancy. PET/MRI is useful in patients with head and neck cancer.