Prediction of Lung Disease using Deep Learning Techniques in Chest X-Ray Images: A Systematic Review

Early detection of lung disorders, such as pneumonia, tuberculosis (TB), and lung cancer (LC), is crucial for effective treatment and improved survival rates. Recent advancements in Deep Learning (DL) algorithms have yielded promising results in medical image analysis, particularly in the diagnosis of lung diseases using chest X-ray (CXR) images. This systematic study provides an overview of various DL architectures and methodologies employed for lung disease prediction using CXR data. The applicability of multiple neural network architectures, including convolutional neural networks (CNNs), transfer learning models, and hybrid approaches, in detecting abnormalities in CXR images is examined. Furthermore, the role of performance evaluation metrics in improving model accuracy is discussed. The study also addresses challenges in clinical adoption, such as dataset quality, model interpretability, and generalizability across diverse populations. Finally, potential research directions are proposed, including the integration of multi-modal data and the application of explainable AI, to enhance the reliability and clinical utility of DL-based lung disease prediction systems in real-world healthcare settings.