Study of Vision Transformers for Covid-19 Detection from Chest X-rays

TL;DR

This paper evaluates various vision transformer models for COVID-19 detection from chest X-rays, demonstrating that they outperform traditional CNNs and achieve near-perfect accuracy, highlighting their potential for clinical diagnosis.

Contribution

The study systematically compares multiple state-of-the-art vision transformer models for COVID-19 detection, showing their superior performance over traditional methods.

Findings

Vision transformers achieved 98.75% to 99.5% accuracy.

Transformers outperformed CNNs and traditional methods.

Transfer learning with ImageNet weights was effective.

Abstract

The COVID-19 pandemic has led to a global health crisis, highlighting the need for rapid and accurate virus detection. This research paper examines transfer learning with vision transformers for COVID-19 detection, known for its excellent performance in image recognition tasks. We leverage the capability of Vision Transformers to capture global context and learn complex patterns from chest X-ray images. In this work, we explored the recent state-of-art transformer models to detect Covid-19 using CXR images such as vision transformer (ViT), Swin-transformer, Max vision transformer (MViT), and Pyramid Vision transformer (PVT). Through the utilization of transfer learning with IMAGENET weights, the models achieved an impressive accuracy range of 98.75% to 99.5%. Our experiments demonstrate that Vision Transformers achieve state-of-the-art performance in COVID-19 detection, outperforming traditional methods and even Convolutional Neural Networks (CNNs). The results highlight the potential of Vision Transformers as a powerful tool for COVID-19 detection, with implications for improving the efficiency and accuracy of screening and diagnosis in clinical settings.