A Comparative Analysis of Colon Cancer Detection using FLC-DO Algorithm

Recently, in classifying and diagnosing various diseases, deep learning algorithms play a major role. There are some deadliest diseases which become a threat to the mankind. So detecting the diseases and curing them earlier is must. Colorectal cancer (the cancer affecting the large intestine) is one amongst them in which it has to be cured earlier to get rid of the risk. The large intestine has to be protected from the tumor because it helps to eat and digest the food that is being consumed. Therefore, early detection of colorectal cancer is very important which helps in preventing mortality rates. Over decades, some Artificial Intelligent (AI) techniques promote the detection, segmentation and classification of the diseases earlier which makes the further treatment easier. This proposed study proves us that how Federated learning with adapted cross-domain oncology classifiers supports in detecting and classifying the colorectal cancer in its earliest stage. Furthermore, the model is compared with the existing models such as ANN, Bi-LSTM and ResNet-50 for better results. The obtained results show that the proposed study has achieved higher accuracy of 98 than any other models. Hence, it proved that the proposed study has an increased efficiency in detecting the colorectal tumors earlier which promotes the disease diagnosis. Keywords: colorectal cancer, artificial intelligence, deep Learning, federated learning with adapted cross-domain oncology classifier.