Drug Efficacy Recommendation System of Glioblastoma (GBM) Using Deep Learning

Glioblastoma Multiforme (GBM) is one of the most aggressive and complex forms of brain cancer, characterized by rapid growth and resistance to most conventional treatments. Personalized medicine, which tailors treatment to individual patients based on genetic data, has become an important focus in cancer research. This project aims to develop a drug efficacy recommendation system for GBM using deep learning techniques. The dataset utilized for this project is sourced from the Genomics of Drug Sensitivity in Cancer (GDSC), which contains genomic and drug sensitivity data across various cancer cell lines. The target variable in the dataset, LN_IC50, is used to represent the sensitivity of cells to different drugs. The system employs multiple deep learning algorithms, including Deep Neural Networks (DNN), Feedforward Neural Networks (FNN), Long Short-Term Memory networks (LSTM), and Capsule Networks (CapsNet), to predict the efficacy of drugs for GBM. The models are evaluated for their ability to predict drug sensitivity accurately, and the results are presented through a user-friendly web application. This research contributes to the field of personalized cancer treatment by providing an efficient way to predict the response of GBM to various drugs, offering a data-driven tool for enhancing decision-making in therapeutic planning.

Keywords: Glioblastoma, Deep Learning, Drug Efficacy, Personalized Medicine, LN_IC50, Cancer Treatment, Genomics, Drug Sensitivity, Predictive Modeling, Machine Learning.