IMPLEMENTASI ALGORITMA CONVOLUTIONAL NEURAL NETWORK UNTUK MENDETEKSI PENYAKIT GINJAL
IMPLEMENTATION OF CONVOLUTIONAL NEURAL NETWORK FOR DETECTING KIDNEY DISEASE
Abstract
There are several types of kidney disease, such as kidney cancer, tumors, etc. Kidney disease can be detected early, to find out what type of disease the patient has. In the world of artificial intelligence, there is a term called Convolutional Neural Network (CNN) which is often used in image data processing. CNN is a category of artificial neural network which is effective in performing image recognition and classification of image data. The purpose of this research is to find out how to apply the CNN algorithm in detecting kidney disease based on existing image data. This research was developed using the Python programming language and will be implemented into a web-based system. The results obtained from this research are the formation of a web-based system, which this website can be used to detect types of kidney disease based on the input images performed. This kidney disease classification website has been successfully created using the Flask Framework with the API from Google Colab which produces the h5 model and Visual Studio Code. Websites can be run on all types of computer operating systems. Image training data using a CNN algorithm derived from 9334 data trains and 3110 data validations. In this case, 4 classes of data image are used, namely cyst kidney data, normal kidney data, tumor kidney data and stone kidney data. It was found that the accuracy of the f1 score was 68%.