Diabetic Retinopathy Detection From Fundus Images Using Multi-Tasking Model With EfficientNet B5

Diabetic Retinopathy (DR) is a common eye disease that affects over 3 million people annually. People with diabetes are more prone to suffer from Retinopathy. This condition can cause blurring of vision and blindness. Early detection treatment the most effective ways manage Due huge number diabetic patients need for accurate automatic diagnosis, development deep neural networks has been acknowledged. One issues learning classification tasks, particularly in medical field, lack labelled training data. Transfer enables model be trained minimal Training Dataset. In Deep convolutional networks, transfer utilized solve problem insufficient Previous studies on have promising. Through color fundus photography study, we developed identify different stages using learning. Our proposed achieved 87% accuracy help EfficientNet model. The main aim this work develop robust system detecting DR automatically.