Efficient pneumonia detection using Vision Transformers on chest X-rays

Abstract

Pneumonia is a widespread and acute respiratory infection that impacts people of all ages. Early detection and treatment of pneumonia are essential for avoiding complications and enhancing clinical results. We can reduce mortality, improve healthcare efficiency, and contribute to the global battle against a disease that has plagued humanity for centuries by devising and deploying effective detection methods. Detecting pneumonia is not only a medical necessity but also a humanitarian imperative and a technological frontier. Chest X-rays are a frequently used imaging modality for diagnosing pneumonia. This paper examines in detail a cutting-edge method for detecting pneumonia implemented on the Vision Transformer (ViT) architecture on a public dataset of chest X-rays available on Kaggle. To acquire global context and spatial relationships from chest X-ray images, the proposed framework deploys the ViT model, which integrates self-attention mechanisms and transformer architecture. According to our experimentation with the proposed Vision Transformer-based framework, it achieves a higher accuracy of 97.61%, sensitivity of 95%, and specificity of 98% in detecting pneumonia from chest X-rays. The ViT model is preferable for capturing global context, comprehending spatial relationships, and processing images that have different resolutions. The framework establishes its efficacy as a robust pneumonia detection solution by surpassing convolutional neural network (CNN) based architectures.

Enhancing pediatric pneumonia diagnosis through masked autoencoders

“…Singh et al fine-tuned the DEIT_B model pretrained on the ImageNet dataset. We replicated the results using the same official data split as ours, yielding an AUC of 0.995 and an accuracy of 94.50% 31 . Nisho et al fine-tuned the EfficientNet with the noisy student network pretrained on the ImageNet dataset.…”

Glaucoma Detection through a Novel Hyperspectral Imaging Band Selection and Vision Transformer Integration

“…However, its application in medical image processing is still modest compared with that in other fields. Several medical imaging research studies have utilized the ViT structure in various domains, including esophageal endoscopic detection, pneumonia detection, MRI imaging, skin cancer detection, and tumor classification [ 44 , 45 , 46 , 47 ]. It is evident that image subjects amenable to ViT application are typically characterized by high spatial information correlation in sequence form, making them well suited for the incorporation of transformer modules such as the self-attention mechanism and position encoding, as shown in Figure 4 .…”