Prediction of Tuberculosis Using Machine Learning Techniques

Abstract

An early and precise diagnosis is necessary to improve treatment outcomes for tuberculosis (TB), which continues to be a major worldwide health concern. This study explores the application of machine learning algorithms to predict tuberculosis from chest X-ray images. By employing advanced techniques like Random Forests, Support Vector Machines (SVM), and Convolutional Neural Networks (CNNs), we created models that could identify TB-related anomalies in X-rays. Data pre-processing techniques like normalization and augmentation improved model performance, and a large dataset of labelled chest X-ray images was used. the proportion of samples that the model accurately separates from samples that are not TB-infected. This was achieved by leveraging the characteristics of True Positives (TP), False Negatives (FN), False Positives (FP), and True Negatives (TN) from the confusion matrix. The models were assessed using various metrics, including accuracy, precision, recall, and F1 score, to ensure their robustness and dependability. Standard performance metrics such as accuracy, recall, precision, F1 score, and false alarm rate were utilized for evaluation, demonstrating the effectiveness of the proposed model, which achieved an accuracy of 93.0%, precision of 92.0%, recall of 94.0%, F1 score of 92.99%, and a false alarm rate of 0.078. Results indicate that machine learning can significantly aid in the early detection of TB, potentially leading to timely interventions. This research highlights the necessity of integrating machine learning tools into clinical practice to improve diagnostic accuracy and ultimately contribute to global TB control efforts.