EfCNN-Net: Smart Detection of Colon and Lung Cancer using Histopathological Images

In low- and middle-income nations where there is little access to health care and cancer screening, the death rate from lung cancer is highest. Contrarily, colon cancer is the subsequent most prevalent cancer worldwide and the third biggest illness overall, accounting for 935,000 fatalities in 2022. The death rate from colon cancer varies by region, with developed nations reporting the highest rates. The odds of survival for lung and colon cancer can be considerably increased with early detection and treatment. Therefore, it’s critical to promote routine cancer screening and prompt treatment while raising awareness of the risk factors linked to these malignancies. Cancer can be found through histopathological image analysis, which makes it possible to spot morphological abnormalities in tissue samples. In this study, we suggest a unique method for using histopathological scans to identify lung and colon tumors. The recommended approach uses convolutional neural networks (CNNs) for feature extraction and categorization. The underlying technology of this method is deep learning. To assess the suggested method and show its efficacy in highly accurate lung and colon cancer detection, we used the LC25000 Lung and colon histopathology imaging collection. On the basis of the parameters and accuracy, we have also contrasted the various image analysis methods. Our findings imply that the EfficientNet algorithm has the potential to increase the precision of cancer detection and support the creation of more efficient therapeutic approaches.