Deep Learning-Based Classification of Lung and Colon Cancer using a Proposed CNN Architecture

Among the main causes of cancer-related death worldwide are lung and colon cancers. Early malignant tumour discovery and precise classification of them define both appropriate treatment and better patient outcomes. This paper provides a convolutional neural network (CNN) model aimed in classifying lung and colon cancer subtypes including colon benign tissue, lung adenocarcinoma, lung benign tissue, and lung squamous cell carcinoma. With 500 images in every class, the model was developed on a 2,500-histology image Kaggle dataset. Since every class gains almost perfect values in accuracy, recall, and F1 scores, our results show a remarkable 99% classification accuracy. To give perfect separation between benign and malignant tissues, the proposed CNN design successfully finds complex patterns in the photos. The study exceeds conventional diagnostic methods since it reveals the great promise of CNNs as a fast, highly accurate non-invasive tool for cancer diagnosis.