Processing and Detection of Lung and Colon Cancer from Histopathological Images using Deep Residual Networks

Lung & Colon cancer are amongst the leading cause of cancer related deaths worldwide. In this study, we used a five-class lung and colon cancer histopathology dataset and applied various image preprocessing techniques such as contrast stretching, unsharp masking, and resizing. We then trained a ResNetl01 model on this preprocessed dataset and achieved a high accuracy of 99.7%. The increased reliability and robustness of deep learning-based histopathology classifiers with high performance computers can enable instantaneous and automated diagnosis which can further help treatment and recovery.