Anika Cheerla

Automated Diagnosis of Alzheimer's

Anika Cheerla, USA Finalist, 13-15

Currently, the diagnosis of Alzheimer’s is a long process, with the outcome largely based on the opinion of the doctor. I wanted to address this issue by creating a tool that quickly and precisely diagnoses Alzheimer’s. In this project, I used neural networks and tested out various structures, algorithms, hidden neurons, and feature sets to find the classifier that would achieve the highest accuracy when determining whether or not a patient has Alzheimer's.