The value of imaging techniques based on enhanced magnetic resonance imaging in the diagnosis and prediction response of radiotherapy for nasopharyngeal carcinoma

Background:The To evaluate the value of imaging techniques based on enhanced magnetic resonance imaging (MRI) in diagnosing and predicting radiotherapy for nasopharyngeal carcinoma (NPC).Materials and Methods: This study analyzed 142 hospitalized patient's clinical data with nasopharyngeal carcinoma from March 2020 to December 2021.The patients were divided into valid and invalid groups.Preoperative T2 weighted image (T2WI) images were used for imaging analysis, feature texture parameters were screened based on R language, and a short-term efficacy prediction model for the training group and the validation group was constructed.Results: A total of 432 different texture parameters were obtained by T2WI image analysis of each patient and 5 characteristic texture parameters were obtained by Least Absolute Shrinkage and Selection Operator (LASSO) dimensionality reduction and 10-fold crossvalidation screening.Specific examples were standard deviation, Cluster Prominence Angle 135 Offset 4, Correlation Angle 135 offset 4, Inertia Angle 135 Offset 4. The prediction models were constructed and the area under the receiver operating characteristic (ROC) curve of the training group model was 0.826 (95% Confidence intervals (CI): 0.708-0.944).The sensitivity and specificity were 83.67% and 69.14%, respectively.The area under the ROC curve of the validation group model was 0.810 (95% CI: 0.682-0.938)and the sensitivity and specificity were 89.46% and 63.29%, respectively.Conclusion: The prediction model constructed has high predictive accuracy, sensitivity and specificity for diagnosing nasopharyngeal carcinoma.The use of enhanced magnetic resonance imaging (MRI)-based imaging to predict the shortterm outcome after radiotherapy for NPC was feasible and the prediction model was stable and reliable.