Cardiovascular disease (CVD) is a leading cause of death globally and 17 people in Singapore die from CVD every day. A doctor has a wide variety of non-invasive and invasive tests at his disposal to diagnose CVD patients. These tests can be expensive and time-consuming for both the patient and the doctor. Therefore, a doctor needs to figure out the most efficient method to predict onset of CVD in his patients.
Our team proposes here an application that is able to quickly diagnose if a patient might contract CVD with a few targeted queries about some clinical measures and lifestyle practices of the patient. The application matches the information provided about the patient with a rule engine that is constructed in consideration to a decision tree model trained using a historical dataset of patients, some of who may have CVD. This application is user-friendly as it guides the user through a dedicated set of queries. The user can quickly derive a first prediction about the patient’s possibility of contracting CVD, which can be a critical and timely reminder to initiate further medical investigations. Such pre-emptive efforts that can be realized with this system may help to save more lives, via an early warning process.
A short questionnaire designed from rules derived from a historical dataset of CVD predictors. A doctor will order a single non-invasive test, the Thallium stress test, and few more non-invasive tests as required by the questionnaire. Based on the rules set by our machine learning algorithm, the questionnaire is able to predict the onset of CVD.
| Official Full Name | Student ID (MTech Applicable) | Work Items (Who Did What) | Email (Optional) |
|---|---|---|---|
| PEE Kian Soon | A0213510B | Overall system design, modelling, system implementation and project management | e0508611@u.nus.edu |
| TANG Shao Qiang | A0213558A | Overall system design, modelling, system implementation and documentation | e0508659@u.nus.edu |
| MAHATHIR Humaidi | A0213456A | Overall system design, modelling, video and github | e0508587@u.nus.edu |
Refer to appendix <Installation & User Guide> in project report at Github Folder: ProjectReport
Supported Browsers
• Firefox 76 • IE 11 • Google Chrome 81
Backend Dependencies (Packages)
The system runs on Django – a Python-based free and open-source web framework. To run the system successfully, some basic packages are required with the latest version of Python 3 (i.e. 3.82).
• pip install Django • pip install django-crispy-forms • pip install experta
Deploy in Local Machine
Open the command prompt terminal and change the working directory to "mysite". Please ensure that all required dependencies as mentioned earlier are successfully installed before proceeding else you will end with build failures.
Run the command "python manage.py runserver" in the command prompt. You should see the following output on the command prompt.
Open a web brower to access the URL "http://127.0.0.1.8000/ to start using the App
Refer to project report at Github Folder: ProjectReport
Refer to Github Folder: Miscellaneous