DOI: 10.14704/NQ.2022.20.11.NQ66032

Android App Development for X-Ray Image Classification using Convolutional Neural Network

Yash G. Bambala, Ronit R. Shahua, Richa R. Khandelwala

Abstract

Automated detection of highly communicable diseases such as COVID-19 will reduce the risk of the disease spreading. Detection of such disease requires a great amount of accuracy to ensure the safety of the patient and doctor as well. To provide an efficient solution to the problem, this model has proposed a robust solution to detect and classify diseases such as Covid-19 and Pneumonia for testing personnel which include doctors, different pathology laboratory personnel, etc. The work is divided into 2 parts, first is image classification into the above mentioned categories using deep neural networks and the second is adding the obtained model for the said deep neural network to an android application using Android Studio. This algorithm classifies the given image into 3 categories namely Covid-19, Pneumonia, and normal. The classifier architecture is created using CNN (Convolutional Neural Network) and it is trained on 10725 images, each of size 128 x 128. This dataset is a collection of datasets taken from Kaggle repositories in their open challenges.

Keywords

pneumonia disease ; X-ray Image; convolutional neural networks; COVID19

Full Text

References