Lung Disease Classification in X-ray Images using a Modified Deep Convolutional Neural Networks

Diagnosing lung diseases is a complex process that requires the integration of clinical data, including test results, medical history, and patient symptoms, with medical imaging modalities such as CT scans and X-rays. The accurate classification of lung diseases plays a pivotal role in healthcare by enabling precise diagnosis and effective treatment of various conditions. Deep learning models have emerged as valuable tools in this domain, leveraging large datasets to improve diagnostic accuracy. Nevertheless, challenges such as overfitting and limited generalization remain prevalent, particularly when working with small datasets. To mitigate these issues, data augmentation techniques are employed to expand the training dataset by generating diverse and realistic variations of the available samples. In this study, data augmentation strategies were applied to evaluate the performance of three pre-trained deep neural networks: VGG-16, DenseNet-121, and DenseNet-201. The comparative analysis revealed that DenseNet-201 outperformed the other models, achieving an accuracy of 83%, precision of 80.25%, recall of 79%, and an F1 score of 78.75%. These results highlight the potential of DenseNet-201 as a robust model for lung disease classification, demonstrating its ability to accurately identify lung diseases and differentiate them from other medical conditions. This study underscores the significance of integrating advanced deep learning methodologies with medical diagnostics, offering promising implications for enhancing disease recognition and improving patient care in clinical settings.