Heuristic Pneumonia and Tuberculosis Detection in X-Ray Images Using Convolutional Neural Networks
Lidia Mekuanint Bantelay, Mesfin Abebe, Rajesh Sharma R, Akey Sungheetha, N. Sengottaiyan
Abstract
Pneumonia and tuberculosis are major public health problems worldwide. These diseases affect the lungs, and if not diagnosed properly in time, they can become fatal. Chest x-ray images are widely used for Pneumonia and TB disease detection and diagnosis. Detection of Pneumonia and TB from chest x-ray images is difficult and needs experience due to similar pathological features of the diseases. Sometimes a misdiagnosis occurs due to this. Several researchers used deep learning and machine learning techniques to solve this problem. These studies used chest x-ray images exclusively to develop Pneumonia and TB disease detection models. But using only chest x-ray images can not necessarily lead to accurate disease detection. Medical records are required to interpret chest x-ray images in the appropriate clinical context. This work develops a multi-input Pneumonia and TB detection model using chest x-ray images and medical records. Convolutional neural network for the chest x-ray image data and Multilayer perceptron for the medical record data are applied to develop the model. We performed feature level concatenation to join the output feature vectors from Convolutional neural network and Multilayer perceptron for disease detection model development. For comparison, we also developed an image-only and medical record-only model. Consequently, the image-only model gives an accuracy of 92.68%, the medical record-only model results in 98.72% accuracy, and the joint model accuracy is recorded at 99.61 %.