Machine learning and deep learning methods for cancer prediction and responses to its treatment

Abstract

Cancer remains a significant global health challenge, affecting individuals of all ages. Early detection and personalized treatment are crucial as they significantly improve prognosis and treatment outcomes. Recent advancements in machine learning (ML) and deep learning (DL) methods, have shown considerable promise in enhancing cancer detection through medical image analysis and predicting patient-specific drug responses. This study focuses on the classi- fication of Gliomas, a type of brain tumor, into Low-Grade Gliomas (LGG) and High-Grade Gliomas (HGG) by proposing an end-to-end tumor grading model that performs on MRI slices. Additionally, it explores the development of a predictive model for cancer drug response by leveraging drug molecular data and clinical cell line information.