🤖 AI Summary
This study addresses binary classification of thoracic diseases from chest X-ray images, leveraging a clinical dataset of 5,824 images. Method: We systematically compare a baseline CNN against DenseNet-121, fine-tuning the latter via transfer learning and integrating Grad-CAM for lesion localization and enhanced decision interpretability. Results: DenseNet-121 significantly outperforms the baseline—achieving a 3.2% higher accuracy and a 0.042 improvement in AUC. Its dense connectivity enables more effective capture of localized pathological features, yielding superior localization precision in pulmonary abnormal regions and greater classification robustness. This work validates DenseNet-121’s efficacy in medical imaging tasks with limited training data and, through rigorous interpretability analysis, provides clinically credible, human-interpretable decision support for AI-assisted diagnosis.
📝 Abstract
Machine learning and artificial intelligence are fast-growing fields of research in which data is used to train algorithms, learn patterns, and make predictions. This approach helps to solve seemingly intricate problems with significant accuracy without explicit programming by recognizing complex relationships in data. Taking an example of 5824 chest X-ray images, we implement two machine learning algorithms, namely, a baseline convolutional neural network (CNN) and a DenseNet-121, and present our analysis in making machine-learned predictions in predicting patients with ailments. Both baseline CNN and DenseNet-121 perform very well in the binary classification problem presented in this work. Gradient-weighted class activation mapping shows that DenseNet-121 correctly focuses on essential parts of the input chest X-ray images in its decision-making more than the baseline CNN.