Early Prediction of Non-Small Cell Lung Cancer and its type using Feature Extraction Algorithm: A Comparison with ANN and SVM

Main Article Content

A. SHAKUL HAMID, M. SHEIK MANSOOR, K. A. MOHAMED RIYAZUDEEN

Abstract

Lung Cancer is one of the most prevalent diseases across the globe. The rate at which the people prone to small and non-small cell lung cancers is very huge when compared with other types such as breast cancer, colon cancer, liver cancer and so on. Doctors face several difficult situations while handling cancer patients. Detection and Prognosis of non-small cell lung cancer at the prior stage helps to cure or to extend the survival rate at different stages. Following the early detection, physicians use either CT or PET images in order to identify the type and the kind of treatment to be considered. Machine learning proves to be the best in eradicating the traditional approaches used for the early prognosis of non-small cell lung cancer. The paper proposes a model to identify the subtype of non-small cell lung cancer with the help of CT images from the Cancer Imaging Archives Dataset. A comparative study has been performed by using the most feasible supervised machine learning algorithms like Artificial Neural Network (ANN) and Support Vector Machine (SVM). Results show that the performed method works better equivalent to the other models. In order to assist the results, performance metrics like accuracy, prediction and recall are also estimated and compared.

Article Details

Section
Articles