• Zulqarnain Ahmed
• Dr. M A Anusuya

With the onset of COVID-19 pandemic that affected millions of lives across the globe and was declared as a medical health emergency by the World Health Organization (WHO), the main challenge that was present was to diagnose the disease correctly so that its effects could be mitigated by suitable medicines. In this paper we take a look at a relatively new method of image classification i.e. Vision Transformer (ViT) [1] introduced by Google AI Labs [2] in 2017 and compare it with the prevalent method of Convolutional Neural Networks (CNNs) in medical imagery. Due to the non-availability of sufficient-size and good-quality chest X-ray image dataset, an effective and accurate classification was a challenge. To deal with these complexities such as the availability of a very-small-sized and imbalanced dataset with image-quality issues, the dataset has been preprocessed in different phases using different techniques to achieve an effective training dataset for the CNN and vision transformer models to attain its best performance. The preprocessing stages of the datasets performed in this study include dataset balancing, and data augmentation. In this work we have used the confusion matrix metrics such as accuracy, precision, recall, specificity, and F1-score to compare the results obtained by the two models. The experimental results have shown the overall accuracy as high as 86.5% for CNN and 96.66% for ViT model which demonstrates the good capability of the proposed ViT model in the current application domain and how vision transformers are impacted by dataset size and training resources.

Convolutional neural networks, vision transformers Keras, Tensorflow, COVID19, pandemic, deep learning, confusion matrix

"DIAGNOSIS OF COVID-19 USING X-RAY IMAGES AND CONVOLUTIONAL NEURAL NETWORK AND VISION TRANSFORMER", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 8, page no.c59-c64, August-2022, Available :http://www.jetir.org/papers/JETIR2208207.pdf

"DIAGNOSIS OF COVID-19 USING X-RAY IMAGES AND CONVOLUTIONAL NEURAL NETWORK AND VISION TRANSFORMER", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 8, page no. ppc59-c64, August-2022, Available at : http://www.jetir.org/papers/JETIR2208207.pdf