Cholesterol Level Detection Based on Iris Recognition Using Convolutional Neural Network Method

High cholesterol levels are one of the indications of heart disease and stroke. Checking cholesterol levels is usually done in the laboratory. However, many people are reluctant to check their cholesterol levels regularly, there are various reasons start from time to economic reasons. This cholesterol level detection application was made to overcome the problems above. This application determines cholesterol levels in a person's body by capturing the image of the iris. If the user has high cholesterol levels, then there is a white-grayish circle on the outer circle of the iris. The machine learning method used to detect cholesterol levels is a convolutional neural network. CNN is a development of multi-layer perceptron (MLPs). CNN is usually used for object detection and classification. This application is expected to help users to routinely check cholesterol levels and be aware of heart disease and stroke.This paper is talking about the performance of CNN to classify an image using android smartphone's camera. we used the pre-trained model called inception-v3 for the CNN architecture with two classes for classification. the hyperparameter in CNN that we've tests were the epoch, learning rate, and batch size. we also did several tests to see the performance of our android application to find the best scenario with lighting, angle, and distance. The accuracy of the system is 97,45%.