Building a Convolutional Neural Network Model for Tuberculosis Detection Using Chest X-Ray Images

Abstract

Background: Tuberculosis (TB) is a highly infectious disease with a high mortality rate if left untreated. Traditional diagnostic methods, like skin tests and sputum smear cultures, are unreliable and time-consuming. Artificial Intelligence (AI) and Deep Learning (DL) can revolutionize healthcare by improving disease diagnosis. This study developed an AI system using Convolutional Neural Network (CNN) to detect TB by analyzing digitalized chest X-ray (CXR) images, which can significantly improve the accuracy and speed of TB diagnosis, leading to better patient outcomes. Methods: A CNN model was developed; it uses a methodology that cuts the edges for analyzing the CXR images for detecting the tuberculosis symptoms in it. A database of chest X-ray images for tuberculosis which was gathered by a team of researchers was used to train the model for detecting tuberculosis. Result: This study uses deep learning to predict tuberculosis using a CNN model with 97% accuracy on CXR images. The patient can be informed about the severity of tuberculosis by the model, which analyzes and checks the tuberculosis symptoms in their CXR image. Conclusion: In summary, the advancement of AI and DL has brought about a significant transformation in the healthcare industry, particularly in the detection, diagnosis, and treatment of diseases. The use of AI and DL in tuberculosis diagnosis has been explored in this study through the development of a CNN model that was trained on chest X-ray images. AI and DL can significantly reduce tuberculosis mortality rates by aiding in early detection.