Implementation of Celiac Disease Detection Using a Website-based Artificial Neural Network Approach

Celiac disease is a rare autoimmune disease that attacks healthy small intestinal tissue due to gluten intolerance. This disease can be diagnosed through intestinal biopsy, endoscopy of the small intestine, and laboratory blood tests. Several previous studies have succeeded in developing a celiac disease detection system using data in the form of images, but this research will use numerical and categorical data from public dataset provider which is Kaggle. The methods used in this research start from data preprocessing, including ordinal encoding, data training-validation-test split, data normalization, and the use of the Artificial Neural Network (ANN) algorithm in developing deep learning models. The ANN architecture configuration consists of an input layer, 3 dense layers with 128, 64, 32 neurons each, a ReLU activation function, 3 dropout layers between each dense layer with a rate of 0.25, and an output layer with a softmax activation function. The model was compiled with an Adam optimizer with a learning rate 0.001, categorical cross-entropy loss function, and accuracy metrics. The results obtained were an accuracy of 97.0651% and a loss of 0.110505, while the validation accuracy was 96.826% and the validation loss was 0.147456. The model is then implemented into a web-based application so that users can easily use it to detect celiac disease based on the data entered into the system.