Analysis of Genetic Mutations Using Nature-Inspired Optimization Methods and Classification Approach

Cancer is one of the highly concerning groups of diseases, with many people being diagnosed yearly. Early detection and prognosis of different types of cancer are critical areas of focus in cancer research. The analysis of genes is essential for early diagnosis, therapy, and recovery from the condition. Individualized medicinal medication therapy plays a crucial role in the treatment of cancer. Personalized medicine is recommended to investigate the genetic profiles of individuals with cancer. Cancer tumors usually have hundreds of genetic alterations. The physical labor involved in using personalized medicine has slowed its adoption for cancer therapy. Each genetic modification requires an expert medical opinion based on published research. The purpose of this study is to use an optimization approach for classification on a dataset to increase the precision of classification models. An analysis of Genetic Mutations in cancer patients using nature-inspired optimization methods like Particle Swarm Optimization (PSO), Bee colon Optimization and Genetic Algorithms is discussed. The mutations are classified using machine learning approach. It is observed that an accuracy of Random Forest classifier with PSO supersites other methods which is 71%. This study is confined to selection of standard dataset whereas the mutations may vary with respect to type of cancer and health history of patient which in turn will impact the accuracy. Authors are working on experimenting the real time datasets as a future research.