The primary goal of this project is to develop a machine learning model that predicts the drug response of Alzheimer’s patients by analyzing genomic datasets and patient-specific medical records. Alzheimer’s disease (AD) is a complex neurodegenerative condition for which treatments often show varied results among patients. Traditional treatment approaches lack personalization and may lead to suboptimal outcomes. This project leverages Support Vector Machines (SVM) and Deep Neural Networks (DNN) to analyze patterns in genetic markers and clinical parameters, aiming to classify patients based on their likely responsiveness to different pharmaceutical interventions. The broader goal is to support personalized medicine by offering targeted treatment recommendations, ultimately improving patient outcomes and reducing trial-and-error in drug administration.
The project follows a structured twelve-week plan. Initially, students will study the foundational concepts of SVM and deep learning, and set up their development environment using tools like Python, Anaconda, or Google Colab. The next phase involves gathering and preprocessing large-scale genomic and clinical datasets relevant to Alzheimer’s.
In the core development phase, students will design, train, and evaluate predictive models using both labeled and unlabeled data. Feature selection, model tuning, and cross-validation will be applied to maximize predictive accuracy. Later stages of the project involve testing the model on unseen data, optimizing performance, finalizing system integration, and preparing documentation. The final output includes a functional prototype and a team presentation. While the system is intended to assist healthcare providers, it is not designed to replace clinical expertise and must be used alongside professional medical advice.
