A video classification method for diagnosing ear diseases using otoscope imaging

Otoscopy, the clinical visual examination of the ear, is essential for diagnosing middle ear infections and other ear diseases, but accurate diagnosis requires a degree of clinical expertise. Automated deep learning methods have been explored to aid the diagnostic process, typically relying on selected otoscope images. However, in real-world scenarios, clinical expert selection is not always feasible, and variations in brightness and color pose challenges. We propose a video classification approach using Video Masked Autoencoder (VideoMAE), leveraging all frames in a video to improve diagnostic accuracy and eliminate the need for human frame selection. Our study developed a deep learning model to classify otoscope videos for seven middle ear conditions. Data augmentation techniques enhanced model generalization, and we resampled videos for balanced representation. The model, trained with 224x224 pixel videos with 16 frames each, optimized parameters to minimize classification errors. Results showed improved classification accuracy with data augmentation and balanced resampling, achieving 92.1%±4.0% accuracy.