AttCDCNet: Attention-enhanced Chest Disease Classification using X-Ray Images
Omar Hesham Khater, Abdullahi Sani Shuaib, Sami Ul Haq, Abdul Jabbar Siddiqui
TL;DR
The paper addresses automatic chest disease classification from chest X-ray images to reduce diagnostic delays and biases. It introduces AttCDCNet, which augments DenseNet121 with attention blocks, depthwise separable convolutions, and focal loss to improve accuracy and efficiency. The approach achieves high performance on multi-class chest diseases (e.g., accuracy up to 94.94%, precision 95.14%, recall 94.53%), with Grad-CAM visualizations to support interpretability. This results in a lighter, more accurate model suitable for potential real-time clinical deployment across diverse datasets.
Abstract
Chest X-rays (X-ray images) have been proven to be effective for the diagnosis of chest diseases, including Pneumonia, Lung Opacity, and COVID-19. However, relying on traditional medical methods for diagnosis from X-ray images is prone to delays and inaccuracies because the medical personnel who evaluate the X-ray images may have preconceived biases. For this reason, researchers have proposed the use of deep learning-based techniques to facilitate the diagnosis process. The preeminent method is the use of sophisticated Convolutional Neural Networks (CNNs). In this paper, we propose a novel detection model named \textbf{AttCDCNet} for the task of X-ray image diagnosis, enhancing the popular DenseNet121 model by adding an attention block to help the model focus on the most relevant regions, using focal loss as a loss function to overcome the imbalance of the dataset problem, and utilizing depth-wise convolution to reduce the parameters to make the model lighter. Through extensive experimental evaluations, the proposed model demonstrates exceptional performance, showing better results than the original DenseNet121. The proposed model achieved an accuracy, precision and recall of 94.94%, 95.14% and 94.53%, respectively, on the COVID-19 Radiography Dataset.