Retinopathy Based Multistage Classification of Diabeties

One of the biggest problems faced in biomedical engineering is the non-invasive assessment of the physiological changes that occur within the human body. Particularly, the detection of the abnormalities in the human eye is very difficult due to the numerous complexities involved in the process. Retinal images can be used to determine the nature of the abnormalities that affect the human eye. Standard disease identification techniques from retinal images mostly involve manual intervention. However, since human observation is extremely prone to error, the success rate of these techniques is quite scarce. Diabetic Retinopathy is one such disease of retina which occurs in people suffering from diabetes. It is a multistage progressing disease namely NDPR and PDR. Micro-aneurysms, haemorrhages and exudates are the anomalous features frequently detected in the retinal images of a person afflicted by diabetic retinopathy. Image processing techniques are applied to pre-process the Fundus image, which is followed by segmentation of anomalies. Feature extraction is done and the features that are detected are used to identify the different stages of diabetic retinopathy using Random Forest classification technique. It is observed that, the proposed algorithm results in approximate classification rate up-to 90%