A Deep Learning-Driven Approach for Detecting Lung and Colon Cancer Using Pre-Trained Neural Networks

Lung and colon cancers are two of the most common and deadliest cancers worldwide, claiming countless lives each year. The ability to diagnose these cancers early can make a big difference in patient survival rates. Moreover, recent advancements in deep learning technology show significant promise in enhancing diagnostic accuracy. In this study, we propose a Computer-Aided Diagnosis (CAD) system that utilizes pre-trained Convolutional Neural Networks (CNNs) to accurately classify histopathological images of lung and colon cancers. We utilized the publicly available LC25000 dataset, which contains 25,000 images, to train and evaluate several well-known deep learning models, such as Visual Geometry Group 16 (VGG16), and Visual Geometry Group19 (VGG19). After fine-tuning model parameters such as learning rate, batch size, and the number of training cycles (epochs), our CNN+VGG19 model achieved an impressive accuracy of 99.04%. These findings demonstrate t he potential of combining CNN architectures with transfer learning to develop more reliable diagnostic tools. Our work contributes to ongoing efforts to improve early cancer detection and offers a helpful tool for doctors in detecting lung and colon cancers.