Objective:To development a deep learning（DL） model based on conventional MRI for automatic segmentation and differential diagnosis of nasopharyngeal carcinoma（NPC） and nasopharyngeal lymphoma（NPL）. Methods:The retrospective study included 142 patients with NPL and 292 patients with NPC who underwent conventional MRI at Renmin Hospital of Wuhan University from June 2012 to February 2023. MRI from 80 patients were manually segmented to train the segmentation model. The automatically segmented regions of interest（ROIs） formed four datasets: T1 weighted images（T1WI）, T2 weighted images（T2WI）, T1 weighted contrast-enhanced images（T1CE）, and a combination of T1WI and T2WI. The ImageNet-pretrained ResNet101 model was fine-tuned for the classification task. Statistical analysis was conducted using SPSS 22.0. The Dice coefficient loss was used to evaluate performance of segmentation task. Diagnostic performance was assessed using receiver operating characteristic（ROC） curves. Gradient-weighted class activation mapping（Grad-CAM） was imported to visualize the model's function. Results:The DICE score of the segmentation model reached 0.876 in the testing set. The AUC values of classification models in testing set were as follows: T1WI: 0.78（95%CI 0.67-0.81）, T2WI: 0.75（95%CI 0.72-0.86）, T1CE: 0.84（95%CI 0.76-0.87）, and T1WI+T2WI: 0.93（95%CI 0.85-0.94）. The AUC values for the two clinicians were 0.77（95%CI 0.72-0.82） for the junior, and 0.84（95%CI 0.80-0.89） for the senior. Grad-CAM analysis revealed that the central region of the tumor was highly correlated with the model's classification decisions, while the correlation was lower in the peripheral regions. Conclusion:The deep learning model performed well in differentiating NPC from NPL based on conventional MRI. The T1WI+T2WI combination model exhibited the best performance. The model can assist in the early diagnosis of NPC and NPL, facilitating timely and standardized treatment, which may improve patient prognosis.
Humans ; Nasopharyngeal Carcinoma/diagnostic imaging* ; Deep Learning ; Magnetic Resonance Imaging ; Retrospective Studies ; Nasopharyngeal Neoplasms/diagnostic imaging* ; Lymphoma/diagnostic imaging* ; Diagnosis, Differential ; ROC Curve ; Male ; Female ; Middle Aged ; Adult

OBJECTIVES: To evaluate the effect of artificial intelligence-assisted compressed sensing (ACS) acceleration on MRI radiomic feature extraction and performance of diagnostic staging models for nasopharyngeal carcinoma (NPC) in comparison with conventional parallel imaging (PI). METHODS: A total of 64 patients with newly diagnosed NPC underwent 3.0T MRI using axial T1-weighted (T1W), T2-weighted (T2W), and contrast-enhanced T1-weighted (CE-T1W) sequences. Both PI and ACS protocols were performed using identical imaging parameters. The total scan time for the 3 sequences in ACS group was 227 s, representing a 30% reduction from 312 s in the PI group. Eighteen first-order and 75 texture features were extracted using Pyradiomics. Intraclass correlation coefficients (ICCs) were calculated to assess the agreement between the two acceleration methods. After feature selection using the least absolute shrinkage and selection operator (LASSO), random forest regression models were constructed to distinguish early-stage (T1 and T2) from advanced-stage (T3 and T4) NPC. The diagnostic performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test. RESULTS: ACS-accelerated images demonstrated good radiomic reproducibility, with 86.0% (240/279) of features showing good agreement (ICC>0.75), with mean ICCs for T1W, T2W and CE-T1W sequences of 0.91±0.09, 0.89±0.13 and 0.88±0.11, respectively. The staging prediction models achieved similar AUCs for ACS and PI (0.89 vs 0.90, P=0.991). CONCLUSIONS The MRI radiomic features extracted using ACS and PI techniques are highly consistent, and the ACS-based model shows comparable diagnostic performance to the PI-based model, but ACS significantly reduces the scan time and provides an efficient and reliable acceleration strategy for radiomics in NPC.
Humans ; Nasopharyngeal Neoplasms/diagnosis* ; Magnetic Resonance Imaging/methods* ; Nasopharyngeal Carcinoma ; Neoplasm Staging ; Artificial Intelligence ; Carcinoma ; Female ; Male ; Middle Aged ; Adult ; Radiomics

OBJECTIVES: To evaluate the performance of different multi-modality fusion models for predicting radiation-induced oral mucositis (RIOM) following radiotherapy in patients with nasopharyngeal carcinoma (NPC). METHODS: We retrospectively collected the data from 198 patients with locally advanced NPC who experienced RIOM following radiotherapy at the Affiliated Tumor Hospital of Guangzhou Medical University from September, 2022 to February, 2023. Based on oral radiation dose-volume parameters and clinical features of NPC, basic classification models were developed using different combinations of feature selection algorithms and classifiers and integrated using a multi-criterion decision-making (MCDM)-based classifier fusion (MCF) strategy and its variant, the H-MCF model. The basic classification models, MCF model, the H-MCF model with a single modality or multiple modalities and other ensemble classifiers were compared for performances for predicting RIOM by assessing the area under the ROC curve (AUC), accuracy, sensitivity, and specificity. RESULTS: The H-MCF model, which integrated multi-modality features, achieved the highest accuracy for predicting severe RIOM with an AUC of 0.883, accuracy of 0.850, sensitivity of 0.933, and specificity of 0.800. CONCLUSIONS Compared with each of the individual classifiers, the multimodal multi-classifier fusion algorithm combining clinical and dosimetric modalities demonstrates superior performance in predicting the incidence of severe RIOM in NPC patients following radiotherapy.
Humans ; Nasopharyngeal Carcinoma/radiotherapy* ; Nasopharyngeal Neoplasms/radiotherapy* ; Retrospective Studies ; Stomatitis/diagnosis* ; Algorithms ; Radiation Injuries/diagnosis* ; Female ; Male ; ROC Curve

PURPOSE: Conditional survival (CS) provides important information on survival for a period of time after diagnosis. Currently, information on CS patterns of patients with nasopharyngeal carcinoma (NPC) is lacking. We aimed to analyze survival rate over time and estimate CS for NPC patients using a national population-based registry. MATERIALS AND METHODS: Patients diagnosed with NPC between 1973 and 2007 with at least 5-year follow-up were identified from the Surveillance Epidemiology End Results registry. Traditional survival rates and crude CS estimateswere calculated using Kaplan-Meier analysis. Risk-adjusted survival curves were plotted from the proportional hazards model using the correct group prognosis method. RESULTS: For 7,713 patients analyzed, adjusted baseline 5-year overall survival improved significantly from 36.0% in patients diagnosed in 1973-1979, 41.7% in 1980-1989, 46.6% in 1990-1999, to 54.7% in 2000-2007 (p < 0.01). CS analysis demonstrated that for every additional year survived, adjusted probability of surviving the next 5 years increased from 66.7% (localized), 54.0% (regional), and 35.3% (distant) at the time of diagnosis, to 83.7% (localized), 75.0% (regional), and 62.2% (distant) for patients who had survived 5 years. Adjusted 5-year CS differed among age, sex, tumor histology, ethnicity, and stage subgroups initially, but converged with time. CONCLUSION: Treatment outcomes of NPC patients have greatly improved over the decades. Increases in CS become more prominent in patients with distant disease than in those with localized or regional disease as patients survive longer. CS provides more dynamic prognostic information for patients who have survived a period of time after diagnosis.
Diagnosis ; Epidemiology* ; Follow-Up Studies ; Humans ; Kaplan-Meier Estimate ; Methods ; Nasopharyngeal Neoplasms ; Prognosis ; Proportional Hazards Models ; SEER Program ; Survival Rate

Biopsy, Fine-Needle ; Biopsy, Needle ; Carcinoma ; diagnosis ; Endoscopic Ultrasound-Guided Fine Needle Aspiration ; Humans ; Nasopharyngeal Carcinoma ; diagnosis ; Nasopharyngeal Neoplasms ; diagnosis

OBJECTIVES: To retrospectively evaluate the patients who underwent nasopharyngeal biopsy with imaging and biopsy results, who have or don’t have symptoms for nasopharyngeal pathology and to determine the ratio of the nasopharyngeal cancer cases and other pathologic conditions. METHODS: In this retrospective study, 983 patients who underwent endoscopic nasopharyngeal biopsy for symptomatic nasopharyngeal lesions were included. All pathological results, benign or malign was recorded and classified due to the patients’ presenting symptoms such as symptomatic for nasopharyngeal pathology or asymptomatic. Computed tomography (CT) or magnetic resonance imaging (MRI) reports were also recorded separately as group A for malignancy or group B for not malignancy. RESULTS: Forty-five (4.6%) of 983 biopsies were malignant. In this group, there is no statistically significant difference between symptomatic and asymptomatic group. For malignant pathologies, the sensitivity of MRI was found 88.2% and CT was 61.5%. CONCLUSION: For early diagnosis of nasopharyngeal cancer, all patients admitted to Ear, Nose and Throat (ENT) referral clinics should be examined endoscopically irrespective of their complaints and suspicious cases should be investigated by imaging especially by MRI. If MRI report clearly indicates Thornwaldt cyst or reactive lymphoid hyperplasia and this result is compatible with endoscopic findings, biopsy may not be necessary. Apart from these cases, all suspected lesions should be biopsied.
Adult* ; Biopsy* ; Ear ; Early Diagnosis ; Humans ; Magnetic Resonance Imaging ; Multidetector Computed Tomography ; Nasopharyngeal Neoplasms ; Nasopharynx* ; Nose ; Pathology ; Pharynx ; Pseudolymphoma ; Referral and Consultation ; Retrospective Studies

PURPOSE: To investigate the predictive role of maximum standardized uptake value (SUVmax) of 2-[18F]fluoro-2-deoxy-D-glucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) in nasopharyngeal cancer patients treated with intensity-modulated radiotherapy (IMRT). MATERIALS AND METHODS: Between October 2006 and April 2016, 53 patients were treated with IMRT in two institutions and their PET/CT at the time of diagnosis was reviewed. The SUVmax of their nasopharyngeal lesions and metastatic lymph nodes (LN) was recorded. IMRT was delivered using helical tomotherapy. All patients except for one were treated with concurrent chemoradiation therapy (CCRT). Correlations between SUVmax and patients’ survival and recurrence were analyzed. RESULTS: At a median follow-up time of 31.5 months (range, 3.4 to 98.7 months), the 3-year overall survival (OS) and disease-free survival (DFS) rates were 83.2% and 77.5%, respectively. In univariate analysis, patients with a higher nodal pre-treatment SUVmax (≥ 13.4) demonstrated significantly lower 3-year OS (93.1% vs. 55.5%; p = 0.003), DFS (92.7% vs. 38.5%; p < 0.001), locoregional recurrence-free survival (100% vs. 50.5%; p < 0.001), and distant metastasis-free survival (100% vs. 69.2%; p = 0.004), respectively. In multivariate analysis, high pre-treatment nodal SUVmax (≥ 13.4) was a negative prognostic factor for OS (hazard ratio [HR], 7.799; 95% confidence interval [CI], 1.506–40.397; p = 0.014) and DFS (HR, 9.392; 95% CI, 1.989–44.339; p = 0.005). CONCLUSIONS: High pre-treatment nodal SUVmax was an independent prognosticator of survival and disease progression in nasopharyngeal carcinoma patients treated with IMRT in our cohort. Therefore, nodal SUVmax may provide important information for identifying patients who require more aggressive treatment.
Cohort Studies ; Diagnosis ; Disease Progression ; Disease-Free Survival ; Electrons ; Fluorodeoxyglucose F18* ; Follow-Up Studies ; Humans ; Lymph Nodes ; Multivariate Analysis ; Nasopharyngeal Neoplasms ; Positron-Emission Tomography and Computed Tomography* ; Radiotherapy, Intensity-Modulated* ; Recurrence

Carcinoma ; complications ; Humans ; Male ; Middle Aged ; Nasopharyngeal Neoplasms ; complications ; Parotid Gland ; pathology ; Parotid Neoplasms ; diagnosis ; secondary

OBJECTIVETo investigate the value of serum globulin levels before treatment in predicting the prognosis of patients with nasopharyngeal carcinoma (NPC).

METHODSA total of 127 patients with non-disseminated NPC were recruited between January, 2009 and December, 2013 at Nanfang Hospital. The pretreatment serum globulin levels were analyzed with the receiver-operating characteristic (ROC) curve analysis to select the cut-off point for low and high pretreatment serum globulin levels. Kaplan-Meier and multivariable analyses were used to evaluate the predictive value of serum globulin levels.

RESULTSThe ROC curve analysis determined 30.05 g/L as the optimal cut-off value for pretreatment serum globulin level, which was significantly associated with gender (P=0.024) and N stage (P=0.016). Kaplan-Meier analysis showed that a high pretreatment serum globulin level (>30.05 g/L) significantly predicted poor progression-free survival (P=0.019), overall survival (P=0.034) and distant metastasis-free survival (P=0.049); multivariate analysis identified pretreatment serum globulin level as an independent prognostic factor for progression-free survival (HR=2.344, P=0.031).

CONCLUSIONPretreatment serum globulin level may serve as a valuable marker to predict the prognosis of patients with NPC.

Carcinoma ; Disease-Free Survival ; Humans ; Kaplan-Meier Estimate ; Multivariate Analysis ; Nasopharyngeal Neoplasms ; blood ; diagnosis ; Prognosis ; ROC Curve ; Serum Globulins ; analysis

Carcinoma ; Humans ; Limbic Encephalitis ; diagnosis ; surgery ; Male ; Middle Aged ; Nasopharyngeal Neoplasms ; diagnosis ; surgery