Al-Qadisiyah University’s Master’s Dissertation: On the Diagnosis of Diabetic Retinopathy Based on Improved Convolutional Neural Network
News of Universities
190
31-01-2024
College of Computer Science and Information Technology, University of Al-Qadisiyah discussed a master dissertation on diagnosing of diabetic retinopathy based on an improved convolutional neural network by the postgraduate student, Mrs. Hoda Abd Al-Hussein Abdel Amir.
The dissertation delivered a study on the field of detection of diabetic retinopathy based on deep learning, with a focus on improving the model’s performance by modifying the hyperparameters.
The dissertation reviewed using the parameters Messidor-2 and APTOS 2019, which are two different data sets used to train the two famous convolutional neural network (CNN) architectures inception-V3 and efficient Net B3.
The dissertation concluded that convolutional neural networks via using the modified Eeficient Net B3 was a highly effective architecture for diagnosing diabetic retinopathy across the five stages of the disease after comparing the results of the two approaches.
