This study aimed to develop deep learning (DL) models for differentiating between eosinophilic chronic rhinosinusitis (ECRS) and non-ECRS (NECRS) on preoperative CT. Axial spiral CT images were pre-processed and used to build the dataset. Two semantic segmentation models based on U-net and Deeplabv3 were trained to segment the sinus area on CT images. All patient images were segmented using the better-performing segmentation model and used for training and testing of the transferred efficientnet_b0, resnet50, inception_resnet_v2, and Xception neural networks. Additionally, we evaluated the performances of the models trained using each image and each patient as a unit. A total of 878 chronic rhinosinusitis (CRS) patients undergoing nasal endoscopic surgery at Renmin Hospital of Wuhan University (Hubei, China) between October 2016 to June 2021 were included. The precision of each model was assessed based on the receiver operating characteristic curve. Further, we analyzed the confusion matrix and accuracy of each model. The Dice coefficients of U-net and Deeplabv3 were 0.953 and 0.961, respectively. The average area under the curve and mean accuracy values of the four networks were 0.848 and 0.762 for models trained using a single image as a unit, while the corresponding values for models trained using each patient as a unit were 0.893 and 0.853, respectively. Combining semantic segmentation with classification networks could effectively distinguish between patients with ECRS and those with NECRS based on preoperative sinus CT images. Furthermore, labeling each patient to build a dataset for classification may be more reliable than labeling each medical image.
Read full abstract