deep learning automated differential diagnosis of pharyngitis using smartphone camera

In this article, we aimed to diagnose whether patients have bacterial pharyngitis ornonbacterial pharyngitis. to achieve this, a dataset from 579 patients was collected, and at least four general practitioners diagnosed each sample. data augmentation methods were employed to increase the sample size, and various preprocessing techniques were applied to enhance the quality of images. in this study, we utilized a convolutional neural network (cnn) and two transformers for binary classification. the results demonstrate the capability of deep learning models to classify pharyngitis into bacterial and nonbacterial categories based on images taken by smartphone cameras with high accuracy.