Alzheimer's Detection
data source: https://www.oasis-brains.org/
The project is to detect Alzheimer's disease using Machine Learning tool on MRI dataset.
We use 5 model of Machine learning (Logistic Regression, AdaBoost, xgboost, Random Forest and Decision Tree) and compare them with each other.
Installation
Create virtual environment
conda create -n Alzheimer python=3.8
conda activate Alzheimer
Install dependencies:
pip install -r requirements.txt
Download and set up data by running
bash setup_data.sh
Usage
Run and save model
python train.py
Expected output:
----------------Result--------------
model F1_score Precision Recall
Make sure folder models
(save best model) exists
Feature
Make sure your input is in exactly in order
Feature | Descripstion |
---|---|
M/F |
Male of Female |
Age |
Age of patient |
EDUC |
Years of education |
SES |
Socioeconomic Status |
MMSE |
Mini Mental State Examination |
eTIV |
Estimated Total Intracranial Volume |
nWBV |
Normalize Whole Brain Volume |
ASF |
Atlas Scaling Factor |
Prediction
To make the prediction
python Alzheime_Detector.py -i *8-Features-above
Each features seperates by comma( , )
The prediction of patient will be Demented or Nondemented
Make prediction using Streamlit API
streamlit run streamlit.py
Then go to the local link and enter patient information
Press Make prediction
button to get the result.