Cognify: Cloud-Based System for Detection of Alzheimer's Disease

Overview

This repository contains all files related to my Final Year Project, which focuses on creating a cloud-based system for detecting Alzheimer's Disease using Deep Learning on MRI Images.

Project Information
Student	Josh Crotty
Supervisor	Bernard Butler

Repository Structure

cognify_application/

Front-end and back-end code for the Cognify web application, built using React and Node.js.

compiling_dataset/

Scripts for processing and structuring raw data obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI).

dataset_eda/

Exploratory data analysis notebooks providing insights into the dataset characteristics.

image_preprocessing/

Collection of scripts implementing various image pre-processing techniques to prepare MRI images for model training.

model_training/

Training scripts organized by model architecture, each in its own subfolder with accompanying notebooks.

model_deployment/

Configuration files and commands used for deploying the model on AWS using TorchServe and Docker.

report/

Final project report created with LaTeX (Overleaf), available in both source and PDF formats.

Technologies Used

• Frontend: React.js
• Backend: Node.js
• Deep Learning: PyTorch, MONAI, Jupyter
• Cloud Services: AWS
• Deployment: Docker, TorchServe
• Documentation: LaTeX

---

© 2025 Josh Crotty | Cloud-Based System for detection of Alzheimer’s Disease using Deep Learning on MRI Images