Criteria for the diagnosis of extranodal extension detected on radiological imaging in head and neck cancer: Head and Neck Cancer International Group consensus recommendations.

Extranodal extension of tumour on histopathology is known to be a negative prognostic factor in head and neck cancer. Compelling evidence suggests that extranodal extension detected on radiological imaging is also a negative prognostic factor. Furthermore, if imaging detected extranodal extension could be identified reliably before the start of treatment, it could be used to guide treatment selection, as patients might be better managed with non-surgical approaches to avoid the toxicity and cost of trimodality therapy (surgery, chemotherapy, and radiotherapy together). There are many aspects of imaging detected extranodal extension that remain unresolved or are without consensus, such as the criteria to best diagnose them and the associated terminology. The Head and Neck Cancer International Group conducted a five-round modified Delphi process with a group of 18 international radiology experts, representing 14 national clinical research groups. We generated consensus recommendations on the terminology and diagnostic criteria for imaging detected extranodal extension to harmonise clinical practice and research. These recommendations have been endorsed by 19 national and international organisations, representing 34 countries. We propose a new classification system to aid diagnosis, which was supported by most of the participating experts over existing systems, and which will require validation in the future. Additionally, we have created an online educational resource for grading imaging detected extranodal extensions.

Utility of Epstein-Barr Virus DNA in Nasopharynx Swabs as a Reflex Test to Triage Seropositive Individuals in Nasopharyngeal Carcinoma Screening Programs.

BACKGROUND: Epstein-Barr virus (EBV) DNA detection in the nasopharynx is considered a biomarker for nasopharyngeal carcinoma (NPC). We evaluated its performance as a reflex test to triage EBV seropositives within an NPC screening program in China. METHODS: The study population was embedded within an ongoing NPC screening trial and included 1111 participants who screened positive for anti-EBV VCA (antibodies against EBV capsid antigens)/EBNA1 (EBV nuclear antigen1)-IgA antibodies (of 18 237 screened). Nasopharynx swabs were collected/tested for EBNA1 gene EBV DNA load. We evaluated performance of EBV DNA in the nasopharynx swab as a reflex test to triage EBV serological high-risk (those referred to endoscopy/MRI) and medium-risk (those referred to accelerated screening) individuals. RESULTS: By the end of 2019, we detected 20 NPC cases from 317 serological high-risk individuals and 4 NPC cases from 794 medium-risk individuals. When used to triage serological high-risk individuals, nasopharynx swab EBV DNA was detected in 19/20 cases (positivity rate among cases: 95.0%; 95% CI, 75.1%-99.9%), with a referral rate of 63.4% (201/317, 95% CI, 57.8%-68.7%) and NPC detection rate among positives of 9.5% (19/201, 95% CI, 5.8%-14.4%). The performance of an algorithm that combined serology with triage of serology high-risk individuals using EBV DNA testing yielded a sensitivity of 72.4% (95% CI, 3.0%-81.4%) and specificity of 97.6% (95% CI, 97.2%-97.9%). When used to triage EBV serological medium-risk individuals, the positivity rate among cases was 75.0% (95% CI, 19.4%-99.4%), with a referral rate of 61.8% (95% CI, 58.4%-65.2%) and NPC detection rate among positives of 0.6% (95% CI, 0.1%-1.8%). CONCLUSIONS: Nasopharynx swab EBV DNA showed promise as a reflex test to triage serology high-risk individuals, reducing referral by ca. 40% with little reduction in sensitivity compared to a serology-only screening program.

Automatic detection and segmentation of morphological changes of the maxillary sinus mucosa on cone-beam computed tomography images using a three-dimensional convolutional neural network.

OBJECTIVES: To propose and evaluate a convolutional neural network (CNN) algorithm for automatic detection and segmentation of mucosal thickening (MT) and mucosal retention cysts (MRCs) in the maxillary sinus on low-dose and full-dose cone-beam computed tomography (CBCT). MATERIALS AND METHODS: A total of 890 maxillary sinuses on 445 CBCT scans were analyzed. The air space, MT, and MRCs in each sinus were manually segmented. Low-dose CBCTs were divided into training, training-monitoring, and testing datasets at a 7:1:2 ratio. Full-dose CBCTs were used as a testing dataset. A three-step CNN algorithm built based on V-Net and support vector regression was trained on low-dose CBCTs and tested on the low-dose and full-dose datasets. Performance for detection of MT and MRCs using area under the curves (AUCs) and for segmentation using Dice similarity coefficient (DSC) was evaluated. RESULTS: For the detection of MT and MRCs, the algorithm achieved AUCs of 0.91 and 0.84 on low-dose scans and of 0.89 and 0.93 on full-dose scans, respectively. The median DSCs for segmenting the air space, MT, and MRCs were 0.972, 0.729, and 0.678 on low-dose scans and 0.968, 0.663, and 0.787 on full-dose scans, respectively. There were no significant differences in the algorithm performance between low-dose and full-dose CBCTs. CONCLUSIONS: The proposed CNN algorithm has the potential to accurately detect and segment MT and MRCs in maxillary sinus on CBCT scans with low-dose and full-dose protocols. CLINICAL RELEVANCE: An implementation of this artificial intelligence application in daily practice as an automated diagnostic and reporting system seems possible.

Comparison of new magnetic resonance imaging grading system with conventional endoscopy for the early detection of nasopharyngeal carcinoma.

BACKGROUND: Although stratifying individuals with respect to nasopharyngeal carcinoma (NPC) risk with Epstein-Barr virus-based markers is possible, the performance of diagnostic methods for detecting lesions among screen-positive individuals is poorly understood. METHODS: The authors prospectively evaluated 882 participants aged 30 to 70 years who were enrolled between October 2014 and November 2018 in an ongoing, population-based NPC screening program and had an elevated NPC risk. Participants were offered endoscopy and magnetic resonance imaging (MRI), and lesions were identified either by biopsy at a follow-up endoscopy or further contact and linkage to the local cancer registry through December 31, 2019. The diagnostic performance characteristics of endoscopy and MRI for NPC detection were investigated. RESULTS: Eighteen of 28 identified NPC cases were detected by both methods, 1 was detected by endoscopy alone, and 9 were detected by MRI alone. MRI had significantly higher sensitivity than endoscopy for NPC detection overall (96.4% vs 67.9%; Pdifference = .021) and for early-stage NPC (95.2% vs 57.1%; P = .021). The sensitivity of endoscopy was suggestively lower among participants who had previously been screened in comparison with those undergoing an initial screening (50.0% vs 81.2%; P = .11). The authors observed a higher overall referral rate by MRI versus endoscopy (17.3% vs 9.1%; P < .001). Cases missed by endoscopy had early-stage disease and were more commonly observed for tumors originating from the pharyngeal recess. CONCLUSIONS: MRI was more sensitive than endoscopy for NPC detection in the context of population screening but required the referral of a higher proportion of screen-positive individuals. The sensitivity of endoscopy was particularly low for individuals who had previously been screened.

Addressing intimate partner violence during the COVID-19 pandemic and beyond: how radiologists can make a difference.

Faced with the COVID-19 pandemic, many countries both in Europe and across the world implemented strict stay-at-home orders. These measures helped to slow the spread of the coronavirus but also led to increased mental and physical health issues for the domestically confined population, including an increase in the occurrence of intimate partner violence (IPV) in many countries. IPV is defined as behavior that inflicts physical, psychological, or sexual harm within an intimate relationship. We believe that as radiologists, we can make a difference by being cognizant of this condition, raising an alert when appropriate and treating suspected victims with care and empathy. The aim of this Special Report is to raise awareness of IPV among radiologists and to suggest strategies by which to identify and support IPV victims. KEY POINTS: • The COVID-19 pandemic led to a marked increase in the number of intimate partner violence (IPV) cases, potentially leading to increased emergency department visits and radiological examinations. • Most IPV-related fractures affect the face, fingers, and upper trunk, and may easily be misinterpreted as routine trauma. • Radiologists should carefully review the medical history of suspicious cases, discuss the suspicion with the referring physician, and proactively engage in a private conversation with the patient, pointing to actionable resources for IPV victims.

Convolutional neural network for discriminating nasopharyngeal carcinoma and benign hyperplasia on MRI.

OBJECTIVES: A convolutional neural network (CNN) was adapted to automatically detect early-stage nasopharyngeal carcinoma (NPC) and discriminate it from benign hyperplasia on a non-contrast-enhanced MRI sequence for potential use in NPC screening programs. METHODS: We retrospectively analyzed 412 patients who underwent T2-weighted MRI, 203 of whom had biopsy-proven primary NPC confined to the nasopharynx (stage T1) and 209 had benign hyperplasia without NPC. Thirteen patients were sampled randomly to monitor the training process. We applied the Residual Attention Network architecture, adapted for three-dimensional MR images, and incorporated a slice-attention mechanism, to produce a CNN score of 0-1 for NPC probability. Threefold cross-validation was performed in 399 patients. CNN scores between the NPC and benign hyperplasia groups were compared using Student's t test. Receiver operating characteristic with the area under the curve (AUC) was performed to identify the optimal CNN score threshold. RESULTS: In each fold, significant differences were observed in the CNN scores between the NPC and benign hyperplasia groups (p < .01). The AUCs ranged from 0.95 to 0.97 with no significant differences between the folds (p = .35 to .92). The combined AUC from all three folds (n = 399) was 0.96, with an optimal CNN score threshold of > 0.71, producing a sensitivity, specificity, and accuracy of 92.4%, 90.6%, and 91.5%, respectively, for NPC detection. CONCLUSION: Our CNN method applied to T2-weighted MRI could discriminate between malignant and benign tissues in the nasopharynx, suggesting that it as a promising approach for the automated detection of early-stage NPC. KEY POINTS: • The convolutional neural network (CNN)-based algorithm could automatically discriminate between malignant and benign diseases using T2-weighted fat-suppressed MR images. • The CNN-based algorithm had an accuracy of 91.5% with an area under the receiver operator characteristic curve of 0.96 for discriminating early-stage T1 nasopharyngeal carcinoma from benign hyperplasia. • The CNN-based algorithm had a sensitivity of 92.4% and specificity of 90.6% for detecting early-stage nasopharyngeal carcinoma.

Pre-treatment amide proton transfer imaging predicts treatment outcome in nasopharyngeal carcinoma.

OBJECTIVE: To investigate the value of pre-treatment amide proton transfer-weighted (APTw) imaging for predicting survival of patients with nasopharyngeal carcinoma (NPC). MATERIALS AND METHODS: Pre-treatment APTw imaging was performed in 77 NPC patients and the mean, 90th percentile, skewness, and kurtosis of APT asymmetry (APTmean, APT90, APTskewness, and APTkurtosis, respectively) were obtained from the primary tumor. Associations of APTw parameters with locoregional relapse-free survival (LRRFS), distant metastasis-free survival (DMFS), and disease-free survival (DFS) after 2 years were assessed by univariable Cox regression analysis and significant APTw parameters, together with age, sex, treatment, and stage as confounding variables, were added to the multivariable model. Kaplan-Meier analysis was used to determine the prognostic significance of patients with high or low APT values based on a threshold value from receiver operating characteristic curve analysis. RESULTS: Locoregional relapse, distant metastases, and disease relapse occurred in 14/77 (18%), 10/77 (13%), and 20/77 (26%) patients, respectively, at a median follow-up of 48.3 (10.6-67.4) months. Univariable analysis showed significant associations of LRRFS with APTskewness (HR = 1.98; p = 0.034), DMFS with APTmean (HR = 2.44; p = 0.033), and APT90 (HR = 1.93; p = 0.009), and DFS with APTmean (HR = 2.01; p = 0.016), APT90 (HR = 1.68; p = 0.009), and APTskewness (HR = 1.85; p = 0.029). In multivariable analysis, the significant predictors for DMFS were APT90 (HR = 3.51; p = 0.004) and nodal stage (HR = 5.95; p = 0.034) and for DFS were APT90 (HR = 1.97; p = 0.010) and age (HR = 0.92; p = 0.014). An APT90 ≥ 4.38% was associated with a significantly poorer DFS at 2 years than APT90 < 4.38% (66% vs. 91%; HR = 4.01; p = 0.005). CONCLUSION: APTw imaging may potentially predict survival in patients with NPC. KEY POINTS: • APTw imaging may provide new markers to predict survival in nasopharyngeal carcinoma. • APT90 is an independent predictor of distant metastases-free survival and disease-free survival. • The APThigh group is at higher risk of disease relapse than the APTlow group.

Intravoxel incoherent motion diffusion-weighted imaging for discrimination of benign and malignant retropharyngeal nodes.

PURPOSE: Anatomical imaging criteria for the diagnosis of malignant head and neck nodes may not always be reliable. This study aimed to evaluate the diagnostic value of conventional diffusion-weighted imaging (DWI) and intravoxel incoherent motion (IVIM) DWI in discriminating benign and malignant metastatic retropharyngeal nodes (RPNs). METHODS: IVIM DWI using 14 b-values was performed on RPNs of 30 patients with newly diagnosed metastatic nasopharyngeal carcinoma (NPC) and 30 patients with elevated plasma Epstein-Barr virus (EBV)-DNA without NPC who were part of an EBV-based NPC screening program. Histogram measurements of the two groups were compared for pure diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion volume fraction (f) and apparent diffusion coefficient (ADC) using the Mann-Whitney U test. Area under the curves (AUCs) of significant measurements were calculated from receiver-operating characteristics analysis and compared using the DeLong test. RESULTS: Compared with metastatic RPNs, benign RPNs had lower ADCmean (0.73 vs 0.82 × 10-3 mm2/s) and Dmean (0.60 vs 0.71 × 10-3 mm2/s) and a higher D*mean (35.21 vs 28.66 × 10-3 mm2/s) (all p < 0.05). There was no difference in the f measurements between the two groups (p = 0.204 to 0.301). Dmean achieved the highest AUC of 0.800, but this was not statistically better than the AUCs of the other parameters (p = 0.148 to 0.991). CONCLUSION: Benign RPNs in patients with EBV-DNA showed greater restriction of diffusion compared with malignant metastatic RPNs from NPC. IVIM did not show a significant advantage over conventional DWI in discriminating benign and malignant nodes.

Distinguishing early-stage nasopharyngeal carcinoma from benign hyperplasia using intravoxel incoherent motion diffusion-weighted MRI.

OBJECTIVES: MRI can detect early-stage nasopharyngeal carcinoma (NPC), but the detection is more challenging in early-stage NPCs because they must be distinguished from benign hyperplasia in the nasopharynx. This study aimed to determine whether intravoxel incoherent motion diffusion-weighted imaging (IVIM DWI) MRI could distinguish between these two entities. METHODS: Thirty-four subjects with early-stage NPC and 30 subjects with benign hyperplasia prospectively underwent IVIM DWI. The mean pure diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion fraction (f) and apparent diffusion coefficient (ADC) values were calculated for all subjects and compared between the 2 groups using Student's t test. Receiver operating characteristics with the area under the curve (AUC) was used to identify the optimal threshold for all significant parameters, and the corresponding diagnostic performance was calculated. A p value of < 0.05 was considered statistically significant. RESULTS: Compared with benign hyperplasia, early-stage NPC exhibited a significantly lower D mean (0.64 ± 0.06 vs 0.87 ± 0.11 × 10-3 mm2/s), ADC0-1000 mean (0.77 ± 0.08 vs 1.00 ± 0.13 × 10-3 mm2/s), ADC300-1000 (0.63 ± 0.05 vs 0.86 ± 0.10 × 10-3 mm2/s) and a higher D* mean (32.66 ± 4.79 vs 21.96 ± 5.21 × 10-3 mm2/s) (all p < 0.001). No significant difference in the f mean was observed between the two groups (p = 0.216). The D and ADC300-1000 mean had the highest AUC of 0.985 and 0.988, respectively, and the D mean of < 0.75 × 10-3 mm2/s yielded the highest sensitivity, specificity and accuracy (100%, 93.3% and 96.9%, respectively) in distinguishing early-stage NPC from benign hyperplasia. CONCLUSION: DWI has potential to distinguish early-stage NPC from benign hyperplasia and D and ADC300-1000 mean were the most promising parameters. KEY POINTS: • Diffusion-weighted imaging has potential to distinguish early-stage nasopharyngeal carcinoma from benign hyperplasia in the nasopharynx. • The pure diffusion coefficient, pseudo-diffusion coefficient from intravoxel incoherent motion model and apparent diffusion coefficient from conventional diffusion-weighted imaging were significant parameters for distinguishing these two entities in the nasopharynx. • The pure diffusion coefficient, followed by apparent diffusion coefficient, may be the most promising parameters to be used in screening studies to help detect early-stage nasopharyngeal carcinoma.

Amide proton transfer MRI detects early changes in nasopharyngeal carcinoma: providing a potential imaging marker for treatment response.

PURPOSE: To determine if treatment of nasopharyngeal carcinoma (NPC) induces early changes in amide proton transfer-weighted (APTw) magnetic resonance imaging (MRI), and to perform a preliminary evaluation of APTw imaging in response assessment. METHODS: Sixteen patients with NPC planned for treatment with radiotherapy and/or chemotherapy underwent APTw imaging of the primary tumour pre-treatment and 2-week intra-treatment. Difference in pre- and intra-treatment APT mean (APTmean) was compared using the Wilcoxon signed rank test. Differences in APTmean and percentage change (%Δ) in APTmean were compared between responders and non-responders based on the outcome at 6 months, using the Mann-Whitney U test. RESULTS: APTmean decreased in 9/16 (56.3%) and increased in 7/16 (43.7%) with no significant difference between the pre- and intra-treatment APT values for the whole group (p > 0.05). NPC showed response in 11/16 (68.8%) and non-response in 5/11 (31.2%). There were significant differences between the %Δ of responders and non-responders for APTmean (p = 0.01). Responders showed %Δ decrease in APTmean of - 23.12% while non-responders showed a %Δ increase in APTmean of + 102.28%. CONCLUSION: APT value changes can be detected in early intra-treatment. Intra-treatment %Δ APTmean shows potential in predicting short-term outcome.

Head and Neck Tumors: Amide Proton Transfer MRI.

Purpose To evaluate the utility of amide proton transfer (APT) imaging in the characterization of head and neck tumors. Materials and Methods This retrospective study of APT imaging included 117 patients with 70 nasopharyngeal undifferentiated carcinomas (NUCs), 26 squamous cell carcinomas (SCCs), eight non-Hodgkin lymphomas (NHLs), and 13 benign salivary gland tumors (BSGTs). Normal tissues were examined in 25 patients. The APT means of malignant tumors, normal tissues, and benign tumors were calculated and compared with the Student t test and analysis of variance. The added value of the mean APT to the mean apparent diffusion coefficient (ADC) for differentiating malignant and benign tumors was evaluated by using receiver operating characteristic analysis and integrated discrimination index. Results The mean APT of malignant tumors (2.40% ± 0.97 [standard deviation]) was significantly higher than that of brain tissue (1.13% ± 0.43), muscle tissue (0.23% ± 0.73), and benign tumors (1.32% ± 1.20) (P < .001). There were no differences between malignant groups (NUC, 2.37% ± 0.90; SCC, 2.41% ± 1.16; NHL, 2.65% ± 0.89; P = .45 to P = .86). The mean ADC of malignant tumors ([0.85 ± 0.17] × 10-3 mm2/sec) was significantly lower than that of benign tumors ([1.46 ± 0.47] × 10-3 mm2/sec) (P = .001). Adding APT to ADC increased the area under the curve from 0.87 to 0.96, with an integrated discrimination index of 7.6% (P = .13). Conclusion These preliminary data demonstrate differences in amide proton transfer (APT) mean of malignant tumors, normal tissues, and benign tumors, although APT mean could not be used to differentiate between malignant tumor groups. APT imaging has the potential to be of added value to apparent diffusion coefficient in differentiating malignant from benign tumors.

Nasopharyngeal carcinoma: relationship between invasion of the prevertebral space and distant metastases.

PURPOSE: To identify primary sites of nasopharyngeal carcinoma (NPC) invasion on the staging head and neck magnetic resonance imaging (MRI) that correlate with distant metastases (DM). MATERIALS AND METHODS: Staging head and neck MRI examinations of 579 NPC patients were assessed for primary tumour invasion into 16 individual sites, primary stage (T) and nodal stage (N). Results were correlated with distant metastasis-free survival (DMFS) using the Cox regression, and the diagnostic performance of significant independent markers for DM was calculated. In addition, sites of primary tumour invasion were correlated also with involvement of the first echelon of ipsilateral nodes (FEN+) using logistic regression. RESULTS: Distant metastases were present in 128/579 NPC patients (22.1%) after intensity-modulated radiotherapy (IMRT)/chemo-IMRT and 5-year DMFS was 78.8%. Prevertebral space invasion (PVS+) and N stage, but not T stage, were independent prognostic markers of DMFS (p = 0.016, < 0.001, and 0.433, respectively). Compared to stage N3, PVS invasion had a higher sensitivity (28.1 vs. 68.8%), but lower specificity (90.5 vs. 47.4%) and accuracy (76.7 vs. 48.9%) for correlating patients with DM. PVS invasion, together with parapharyngeal fat space invasion (PPFS+), was also an independent predictive marker of FEN+. CONCLUSION: PVS was the only site of primary tumour invasion that independently correlated with DM, and together with PPFS + was an independent prognostic marker of FEN+, but the low specificity and accuracy of PVS invasion limits its use as a prognostic marker of DM.

MRI of diffuse large B-cell non-Hodgkin's lymphoma of the head and neck: comparison of Waldeyer's ring and sinonasal lymphoma.

To document the magnetic resonance imaging (MRI) features of diffuse large B-cell lymphoma (DLBCL) in Waldeyer's ring (WR) and the sinonasal (SN) region, and to identify any differences between lymphatic and extra-lymphatic DLBCLs, and predictors of disease beyond the neck. Primary, nodal, and multifocal sites on head and neck MRI were compared between 31 WR and 15 SN DLBCL, and between 27 patients with disease confined to the head and neck and 16 patients with disease beyond the neck, using logistic regression. Compared to SN, WR DLBCLs had significantly smaller primary tumour volumes (p = 0.009), less deep invasion (p = 0.001), and more nodal disease (p = 0.016). Tumour site (WR vs. SN) was an independent predictor of deep invasion (p = 0.007). Nodal and multifocal diseases were predictors of disease beyond the neck (p = 0.027 and 0.011, respectively). Lymphatic WR DLBCLs were less locally aggressive but had greater propensity to nodal spread than extra-lymphatic SN DLBCLs. Nodal and multifocal diseases predicted disease beyond the neck.

Diffusion-weighted imaging of nasopharyngeal carcinoma to predict distant metastases.

Our study aimed to identify diffusion-weighted imaging (DWI) parameters obtained from primary nasopharyngeal carcinoma (NPC) at initial presentation, that can predict patients at risk of distant metastases. One hundred and sixty-four patients underwent pretreatment magnetic resonance imaging and DWI. The apparent diffusion coefficient (ADC)mean, ADCskewness, and ADCkurtosis were obtained by histogram analysis. Univariate and multivariate analyses of these ADC parameters together with primary volume (PV), nodal volume (NV), T stage, N stage and presence of locoregional relapse (LRR) were compared between patients with distant metastases (DM+) and patients without distant metastases (DM-) at 5 years using logistic regression. Twenty-eight out of 164 patients (17.1 %) were DM+ (2.5-60 months) and 136/164 patients were DM- (61.2-119.4 months). Compared to DM- patients, the primary tumour of DM+ patients showed significantly lower ADCskewness (ADC values with the greatest frequency were higher) (p = 0.041), and higher PV (p = 0.022), NV (p < 0.01), T stage (p = 0.023), N stage (p < 0.01) and LRR (p < 0.01). On multivariate analysis the ADCskewness was no longer significant (p = 0.120) and only NV and LRR were independent predictors for DM+ (p = 0.023 and 0.021, respectively). DWI showed that compared to DM- patients, DM+ patients had a significantly lower primary tumour ADCskewness, but at initial presentation NV was the only independent predictor of DM.

Narrow band imaging endoscopy of the nasopharynx is not more useful than white light endoscopy for suspected nasopharyngeal carcinoma.

Endoscopy is often used to screen for nasopharyngeal carcinoma. A normal nasopharynx on white light endoscopy may yet harbor subclinical or occult malignancy. This study assessed whether the vascular pattern seen on narrow band imaging endoscopy could indicate this and thus be useful for detecting suspected nasopharyngeal carcinoma. The nasopharynx of 156 patients who failed serological screening for or presented with symptoms of nasopharyngeal carcinoma was graded under white light and narrow band imaging endoscopy and a biopsy taken. The accuracy of assessing the nasopharynx as being probably or definitely malignant on white light endoscopy was high (area under the curve = 0.924), as it was of being normal on narrow band imaging endoscopy (=0.799). The sensitivity and specificity of white light and narrow band imaging endoscopy for nasopharyngeal carcinoma was 93 and 22 %, and 92 and 98 %, respectively. Significantly associated with nasopharyngeal carcinoma was a high index of suspicion or definitely malignant grade on white light endoscopy (p < 0.0005, odds 58.978) and vascular tufts on narrow band imaging endoscopy (p = 0.020, odds 41.210). Narrow band imaging endoscopy of vasculature alone for suspected nasopharyngeal carcinoma is not more useful than white light endoscopy of nasopharyngeal morphology, nor does it add to or surpass the diagnostic accuracy of white light endoscopy in this regard.

Amide proton transfer-weighted imaging of the head and neck at 3 T: a feasibility study on healthy human subjects and patients with head and neck cancer.

The aim of this study was to explore the feasibility and repeatability of amide proton transfer-weighted (APTw) MRI for the head and neck on clinical MRI scanners. Six healthy volunteers and four patients with head and neck tumors underwent APTw MRI scanning at 3 T. The APTw signal was quantified by the asymmetric magnetization transfer ratio (MTRasym) at 3.5 ppm. Z spectra of normal tissues in the head and neck (masseter muscle, parotid glands, submandibular glands and thyroid glands) were analyzed in healthy volunteers. Inter-scan repeatability of APTw MRI was evaluated in six healthy volunteers. Z spectra of patients with head and neck tumors were produced and APTw signals in these tumors were analyzed. APTw MRI scanning was successful for all 10 subjects. The parotid glands showed the highest APTw signal (~7.6% average), whereas the APTw signals in other tissues were relatively moderate. The repeatability of APTw signals from the masseter muscle, parotid gland, submandibular gland and thyroid gland of healthy volunteers was established. Four head and neck tumors showed positive mean APTw ranging from 1.2% to 3.2%, distinguishable from surrounding normal tissues. APTw MRI was feasible for use in the head and neck regions at 3 T. The preliminary results on patients with head and neck tumors indicated the potential of APTw MRI for clinical applications.

Automatic detection of arterial input function in dynamic contrast enhanced MRI based on affinity propagation clustering.

PURPOSE: To automatically and robustly detect the arterial input function (AIF) with high detection accuracy and low computational cost in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). MATERIALS AND METHODS: In this study, we developed an automatic AIF detection method using an accelerated version (Fast-AP) of affinity propagation (AP) clustering. The validity of this Fast-AP-based method was proved on two DCE-MRI datasets, i.e., rat kidney and human head and neck. The detailed AIF detection performance of this proposed method was assessed in comparison with other clustering-based methods, namely original AP and K-means, as well as the manual AIF detection method. RESULTS: Both the automatic AP- and Fast-AP-based methods achieved satisfactory AIF detection accuracy, but the computational cost of Fast-AP could be reduced by 64.37-92.10% on rat dataset and 73.18-90.18% on human dataset compared with the cost of AP. The K-means yielded the lowest computational cost, but resulted in the lowest AIF detection accuracy. The experimental results demonstrated that both the AP- and Fast-AP-based methods were insensitive to the initialization of cluster centers, and had superior robustness compared with K-means method. CONCLUSION: The Fast-AP-based method enables automatic AIF detection with high accuracy and efficiency.

Deep learning for the automatic detection and segmentation of parotid gland tumors on MRI.

OBJECTIVES: Parotid gland tumors (PGTs) often occur as incidental findings on magnetic resonance images (MRI) that may be overlooked. This study aimed to construct and validate a deep learning model to automatically identify parotid glands (PGs) with a PGT from normal PGs, and in those with a PGT to segment the tumor. MATERIALS AND METHODS: The nnUNet combined with a PG-specific post-processing procedure was used to develop the deep learning model trained on T1-weighed images (T1WI) in 311 patients (180 PGs with tumors and 442 normal PGs) and fat-suppressed (FS)-T2WI in 257 patients (125 PGs with tumors and 389 normal PGs), for detecting and segmenting PGTs with five-fold cross-validation. Additional validation set separated by time, comprising T1WI in 34 and FS-T2WI in 41 patients, was used to validate the model performance. RESULTS AND CONCLUSION: To identify PGs with tumors from normal PGs, using combined T1WI and FS-T2WI, the deep learning model achieved an accuracy, sensitivity and specificity of 98.2% (497/506), 100% (119/119) and 97.7% (378/387), respectively, in the cross-validation set and 98.5% (67/68), 100% (20/20) and 97.9% (47/48), respectively, in the validation set. For patients with PGTs, automatic segmentation of PGTs on T1WI and FS-T2WI achieved mean dice coefficients of 86.1% and 84.2%, respectively, in the cross-validation set, and of 85.9% and 81.0%, respectively, in the validation set. The proposed deep learning model may assist the detection and segmentation of PGTs and, by acting as a second pair of eyes, ensure that incidentally detected PGTs on MRI are not missed.

Radiologic extranodal extension for nodal staging in nasopharyngeal carcinoma.

PURPOSE: Extranodal extension (ENE) has the potential to add value to the current nodal staging system (N8th) for predicting outcome in nasopharyngeal carcinoma (NPC). This study aimed to incorporate ENE, as well as cervical nodal necrosis (CNN) to the current stage N3 and evaluated their impact on outcome prediction. The findings were validated on an external cohort. METHODS & MATERIALS: Pre-treatment MRI of 750 patients from the internal cohort were retrospectively reviewed. Predictive values of six modified nodal staging systems that incorporated four patterns of ENE and two patterns of CNN to the current stage N3 for disease-free survival (DFS) were compared with that of N8th using multivariate cox-regression and concordance statistics in the internal cohort. Performance of stage N3 for predicting disease recurrence was calculated. An external cohort of 179 patients was used to validate the findings. RESULTS: Incorporation of advanced ENE, which infiltrates into adjacent muscle/skin/salivary glands outperformed the other five modifications for predicting outcomes (p < 0.01) and achieved a significantly higher c-index for 5-year DFS (0.69 vs 0.72) (p < 0.01) when compared with that of N8th staging system. By adding advanced ENE to the current N3 increased the sensitivity for predicting disease recurrence from 22.4 % to 47.1 %. The finding was validated in the external cohort (5-year DFS 0.65 vs. 0.72, p < 0.01; sensitivity of stage N3 increased from 14.0 % to 41.9 % for disease recurrence). CONCLUSION: Results from two centre cohorts confirmed that the radiological advanced ENE should be considered as a criterion for stage N3 disease in NPC.