Early Alzheimer's Risk Prediction

As one of the most prevalent progressive neural degenerative disease, Alzheimer's is one of the top factors that affects the quality of life of the elderly population.

One of the key methods to combat Alzheimer's is an early discovery, but that usually entails a costly and lengthy process. To help address that problem, my team and I designed a decision-tree-based machine learning algorithm to help clinicians identify people with high Alzheimer's risks. The evaluation can be done with cheap and accessible data, which highlights the value of the solutioh.

My main role in this project is to explore the boosting algorithm family. A fine-tuned gradient boosting model that I built eventually was adopted as the champion model for this project.

Feel free to check out the entire project on GitHub.

Download Project