Identification of Cholesterol Disease Symptoms Based on Iris Image Using Artificial Neural Networks and Backpropagation Method
Cholesterol is a fat that is in human blood for the formation of several hormones and new cell walls. However, the recommended level of cholesterol in the blood for each person varies. It depends on each person whether they have a higher or lower risk of being diagnosed as a disease. Normal human cholesterol levels are in the range of 200 mg/dL or less. The purpose of this study is to create an identification system that can identify symptoms of cholesterol disease based on the image of the iris of the eye using an artificial neural network with the backpropagation method. This study collected training data of 196 iris data consisting of 70 cholesterol iris data, 70 normal iris data, and 56 test iris data consisting of 28 cholesterol iris data and 28 normal iris data. From this study, the best accuracy results were obtained using a parameter with a value of Learning Rate = 0.1 in the 100th iteration resulting in the highest percentage of 89.28% occurring in training data, while the test data yielding the highest percentage of 87.50%.