The way patients engage with medical professionals for treatment has altered as a result of smart healthcare. However, with such intelligent automated systems, security and support for different diseases remain issues. Diabetic Retinopathy (DR), one of the serious conditions, is a serious worry for those who have had diabetes for a long time and can cause total blindness in people of any age. Additionally, block chain has become more well-known in recent years for enabling secure sender-receiver communication.Therefore, the goal of this effort is to develop a blockchain-based intelligent healthcare system for diabetic retinopathy early detection. However, early DR discovery comes with complications and necessitates a professional diagnosis,not accessible anywhere. As a result, the suggested smart healthcare paradigm includes computer-aided diagnosis (CAD) support for early disease symptom identification. The early diagnosis of DR, which necessitates extensive research to create an accurate and efficient model that can function without human contact, may be aided by the CAD model. In order to create the optimal model for DR early detection, this paper offers an empirical investigation of these variables. IoT-based smart devices that can identify diabetic retinopathy (DR) in patients can be developed using the best model. The report also discusses the significance of blockchain-based technologies and the Internet of Things for the creation of intelligent healthcare systems. A suggested 2-branch CNN model makes use of the parameters' values and the kind of hyper parameters selected from the study and the Kaggle fundus image set is used to validate the model. The suggested 2-branch CNN model performs exceptionally well when different parameters are analyzed and their optimal values are used.

medical diagnosis, fundus pictures, diabetic retinopathy, healthcare system, 2-branch CNN, and internet of things