Hyper parameter tuning based gradient boosting algorithm for detection of diabetic retinopathy: an analytical review

The pipelines of approaches for classifying diabetic retinopathy were examined in this study. effort entails developing appropriate transformations and estimators that can be used to automate the process detection. segmentation blood vessels was done using a hybrid algorithm uses Otsu median filter get region interest. Further, ten classifiers investigated order develop an automated pipeline reviewed based on earlier work similar setting exploration new ways identifying retinopathy. To overcome challenge low volume dataset, data argumentation so generic classifier configured. Extensive hyper parameter tuning performed, it shown gradient boosting approach is most stable technique detecting This validated 10K fold cross validation method many metrics (accuracy, recall, precision, v-measure score). Hyper-parameter helped achieving accuracy 0.96.