Vision Transformer based ResNet Model for Pneumonia Prediction

This research study aims to develop a pneumonia detection system using vision transformers. Pneumonia is a very serious respiratory illness that may result in severe health issues, and early detection is essential for effective treatment. Deep learning-based computer vision algorithms have yielded encouraging results in medical image analysis in recent years, and vision transformers have emerged as a potent tool for processing visual data. The proposed system uses a vision transformer to process chest x-ray images and extract visual traits, which can be utilized for classification. The proposed model "Vit_base_resnet50_224_in21k" is trained on a vast and diverse dataset of annotated chest x-ray images to understand the patterns and characteristics of pneumonia. The system's performance is estimated using standard evaluation parameters - accuracy, loss, specificity, sensitivity, F1-score and ROC curve. The proposed model has achieved a performance accuracy of 92.96% for pneumonia detection. The results demonstrate the potential of vision transformers in chest x-ray image analysis and contribute to the development of more accurate and efficient tools for pneumonia detection. This system has the potential to assist healthcare professionals in making faster and more accurate diagnosis, which can ultimately improve outcomes and save lives.