Alzheimer's Brain Scan Analysis
Leveraging Machine Learning to detect subtle neuroanatomical deteriorations in MRI scans, aiding in the early diagnosis of Alzheimer's disease.
🧐 The Challenge
Alzheimer's disease causes physical changes in the brain long before distinct symptoms appear. Detecting these minute structural changes in MRI scans is difficult for the human eye, often leading to missed opportunities for early intervention.
💡 Analysis & Model
This project utilised a Convolutional Neural Network (CNN) architecture. The model was trained on a large dataset of brain MRI scans, categorised by the stage of cognitive impairment. We implemented image pre-processing techniques to normalise the scans for better model performance.
- ✓Image pre-processing: Skull stripping and normalisation.
- ✓3D CNNs to capture volumetric data from MRI.
- ✓High sensitivity in detecting mild cognitive impairment (MCI).
Results & Impact
The model demonstrated a significant ability to differentiate between healthy brains and those with early-stage Alzheimer's. This tool could serve as a "second opinion" for radiologists, increasing diagnostic confidence and speed.
