LungVision in Humans-An Ensemble Method of Classification of Cancer in Lungs

The largest cause of cancer-related fatalities world- wide is lung cancer. Lung cancer early identification is essential for enhancing patient outcomes. Vision Transformers (ViT) and Convolutional Neural Networks (CNN) are examples of deep learning approaches that have recently demonstrated promise in the detection of lung cancer from medical photographs. CNNs are a particular kind of deep neural network that can directly learn features from images. They have been utilized effectively for a number of imaging process in medical applications. On the other hand, ViTs are more recent kind of deep-learning model that has demonstrated success in image categorization tasks. In this paper, a novel strategy for lung cancer classification that ensembles Vision Transformer and CNN models using the dataset of lung Computed Tomography (CT) images that is publicly available in Kaggle. The suggested technique has an accuracy of 98.24% and enhanced lung cancer classification precision and effectiveness, thereby improving patient outcomes.