Publication

Abstract : Cancer is the second-leading cause of death in the world. Uncontrolled cell division of damaged cells causes lung cancer, leading to the formation of tumors. Lung cancer typically originates in the lungs, often in the bronchi or bronchioles. Among various cancer types, lung cancer stands out as one of the most dangerous and deadly forms. However, early diagnosis can significantly improve the patient survival rate. Computed Tomography (CT) is widely regarded as one of the best imaging techniques in the medical field for diagnosis. Nevertheless, in-terpreting CT images accurately, especially when cancer tumors are very small, can be challenging. This is where deep learning models come into play. Deep learning models, a subset of machine learning, play a vital role in interpreting these images for diagnosis. They excel at automatically analyzing vast amounts of CT scan data, enabling the early detection of lung cancer and assisting medical professionals in distinguishing between benign and malignant lesions, thereby contributing to improved patient outcomes. In this study, we have implemented different CNN versions of architectures of Efficientnet and MobileNet. We compared the results of all the models. The models are evaluated based on metricslike accuracy, F1-Score and recall.