med-api
A RESTful API for medical applications of machine learning.
Installation Instructions
- Clone the repository wherever you'd like.
- Create a virtual environment in the 'med-api' directory. For example:
python3 -m virtualenv med-api
And then in the med-api directory:
source bin/activate
- In the med-api directory, install the requirements with pip:
pip3 install -r requirements.txt
- Run app.py
python3 app.py
The app should run. You can test it out with the two test json files. Currently the only feature is giving a prediction for breast cancer:
curl -X POST -H "Content-Type: application/json" -d @test.json http://localhost:3000/api/breast-cancer/prediction
{
"confidence": 0.9999998999999902,
"malignant_status": 0
}
curl -X POST -H "Content-Type: application/json" -d @test2.json http://localhost:3000/api/breast-cancer/prediction
{
"confidence": 0.9999994043330815,
"malignant_status": 0
}
TODO:
- Implementing more classification features.
- Add support for databases.
DISCLAIMER !!
The words and other content provided in this project, and in any linked materials, are not intended and should not be construed as medical advice. If the reader or any other person has a medical concern, he or she should consult with an appropriately-licensed physician or other health care worker.
Never disregard professional medical advice or delay in seeking it because of something you garnered from this project or in any linked materials. If you think you may have a medical emergency, call your doctor immediately.
The views expressed on this project have no relation to those of any academic, hospital, practice or other institution with which the authors are affiliated.