Using Machine Learning Algorithms to Detect Cancer Automatically

Abstract

The number of people diagnosed with cancer is growing all around the world. During the last twenty years, the overall cancer incidence in Iraq has doubled, leading to an increase in the number of diagnosed cancer fatalities. When it comes to deaths that occur in hospitals, cancer is the second-biggest cause. Therefore, a remedy to the issue should be an arrangement to decrease time waste, the right technique of directing the patient to notice symptoms, extremely accurate cancer detection, and a better monitoring system. The proposed method is an arrangement that lets and leads a patient to identify symptoms on their own, guiding them to a proper healthcare professional, correctly diagnosing cancer in its initial stages, and monitoring the patient throughout therapy. Currently, research into cancer detection systems only employs a single machine learning approach to identify cancer. The study that is being presented makes use of Convolutional Neural Networks (CNN), Random Forest, and the XGBoost Classifier, which are a machine learning algorithms that are applied to structured and tabular data in order to identify the existence of breast cancer, brain tumors, skin cancer, and lung cancer. These methods provide findings more quickly while also achieving a greater level of accuracy. Hosting this suggested solution in the cloud with a cutting-edge program will make it available to the public, providing an improved user experience and easier operation.