Classification of lung cancer histology using CT images based on convolutional neural network-CNN

In the field of medical imaging, interest in deep learning as a talented new discipline and a key component is growing rapidly. In lung cancer, the histology of tumours is a key predictor of therapy response and outcome. The most accurate way to classify histologies is through a pathologist's examination of tissue samples. Early tumour detection and therapy are essential for patients' recovery. For the purpose of identifying illness signs and risk stratification, recent developments in machine learning, medical image processing, and the use of radiologic data are highlighted. Regarding lung cancer and its various subtypes, it is error-prone and time-consuming to identify and differentiate. Convolutional Neural Networks can more accurately and quickly diagnose different types of lung cancer, which is essential for choosing the best treatment plan and patient survival rate. This research is concerned with both malignant (tumour) and benign tissue cell carcinomas. A training accuracy of 98.60 percent is obtained for the classification system of the suggested system, which is superior to that of both classic neural networks and the CNN model.