1. Introduction
2. Input dataset
3. Methods
4. System requirements
5. Run instruction
6. Questions?

The goal of this work is to predict the lifetime dementia risk for people. In this project, I used people's demographics, family and healthy history, genetic and behavioral assessment data to build a predictive model using random forest machine learning technique, which not just can help users predict their lifetime dementia risk, but also help them assess the top risk factors for this disease.

The data I used is OASIS-3, collected by Washington university over 15 years. OASIS-3 is the latest release in the Open Access Series of Imaging Studies (OASIS) that aimed at making neuroimaging datasets freely available to the scientific community. OASIS-3 is a longitudinal neuroimaging, clinical, cognitive, and biomarker dataset for normal aging and Alzheimer’s Disease. (See more detail using the link: https://www.oasis-brains.org/).

The input files used in this work are different csv files, residing in the top-most raw data directory of the repository.

The procedure I used to predict the lifetime dementia risk for users are as follows:

1. Preprocess the data, including processing the missing and abnormal values, converting the categorial data to numerical numbers, and merge and match the data from different sources, and etc.
2. Feature selection using correlation heatmap, variance analysis, and univariate linear regression test.
3. PCA analysis;
4. Data augmentation test;
5. Design machine learning pipeline including data normalization, data splitting, model selection, and hyper-parameters tuning;
6. Model evaluation and result analysis;
7. Application design and deployment.

• Both Linux and Windows are supported.
• 64-bit Python 3.7 installation.
• Packages required: Pandas, Numpy, Matplotlib, Sklearn, Seaborn, Streamlit, and Pickle.

1. Run UDS_preprocess_1.py and UDS_preprocess_2.py to preprocess the data;
2. Run data_wrangling.py to merge the data from different tables and do some preliminary analysis on the data;
3. Run exploratory_data_analysis.py to perform the tasks below: build correlation heat map, remove features with low variance, and remove the features that have a low chance to have an effect on the dependent variable.
4. Run predictive_model.py to train the machine learning models and build the predictive model;
5. Run dementia_risk_predictor.py to build the application of predicting the lifetime dementia risk for users.

For any questions, concerns, and comments, please contact Lian Jiang at jiang2015leon@gmail.com