Cholesterol Detection Based on Eyelid Recognition Using Convolutional Neural Network Method

Rizki Mulia Pratama, Astri Novianty, Casi Setianingsih


Lack of public awareness of health will cause serious problems. A small example, people now tend to always consume fatty foods without thinking about the risk of cholesterol levels in the body.  Information on the level of cholesterol suffered by humans can be seen on the human eyelids. The eyelids, one part of the eye, can be known as a person's cholesterol level by observing the eyelids' shape and condition, but many people do not know about this. This application is an application made to detect cholesterol based on the shape of the eyelids. This can determine whether a person is exposed to cholesterol or not, using the Convolutional Neural Network (CNN) method in the classification process. This study provides an output in the form of early detection of cholesterol and prevention so that users can minimize the possibility of illness that will be suffered. This research was conducted to detect cholesterol one eyelid based on digital images. For detecting a cholesterol level, this system got 95.83% of accuracy.


convolutional neural network, CNN, transfer learning method, cholesterol

Full Text: PDF


  • There are currently no refbacks.