Enhancing Deep Learning Performance for Diabetic Retinopathy Diagnosis Through Data Augmentation

Key Points

• This research aims to improve the performance of deep learning algorithms in diagnosing diabetic retinopathy using data augmentation techniques.
• Utilized a deep learning model for diabetic retinopathy diagnosis.
• Applied various data augmentation strategies to increase the diversity of training images.
• Performed comparative analysis to assess the model's performance pre- and post-augmentation.
• Deep learning model accuracy increased from 85% to 92% after applying data augmentation methods.
• Sensitivity improved significantly from 80% to 88%, with a p-value of 0.002.
• Specificity remained consistently high at 95% across both conditions.

Abstract