An Extensible CNN Model for Lung Cancer Diagnosis Using CT Images

Deep learning methods, which form the backbone of neural network architectures, have not only demonstrated exceptional capabilities in classifying data but also in reducing false predictions when handling vast datasets. In the context of clinical applications, particularly those requiring high-stakes decision-making, it is imperative to adopt methodologies that ensure both accurate predictions and precise classifications. Addressing this critical need, our research focused on the development and evaluation of a novel deep-learning framework for the prediction and classification of lung cancer using an extensive dataset comprising approximately 200,000 CT scan images. We proposed an Extensional Convolutional Neural Network (ECNN) model designed to optimize both predictive performance and classification accuracy. The results of our approach were highly promising, achieving a peak accuracy of 97.89%, significantly outperforming existing models in this domain. This substantial improvement underscores the potential of the ECNN model to enhance diagnostic accuracy, thereby contributing to more reliable clinical decision-making in the detection and management of lung cancer.