Diagnosing patients via symptoms using machine learning

Inspiration

My inspiration was from phreesia's speech at the beginning of the

What it does

it is an app that allows users to input symptoms and their age and gender and it then shows them what illness they may have.

How we built it

1.The models used scikit-learn as well as the accuracy metrics. 2.Data prep and manipulation used pandas 3.Front end presentation used Dash

Challenges we ran into

The number of models and the optimization of models had to be cut short due to time constraints. The TSNE couldnt have been implemented.

Accomplishments that we're proud of

The project still has a lot of work to be done, but the completeness of the project is great. The idea has been fully fleshed out in such a short time with only one person and now can be improved.

What we learned

That healthcare is ripe for opportunities with machine learning and just how practical it can be.

What's next for Diagnosing patients via symptoms using machine learning

1.Building a suite of models and leveraging each model in an ensemble model. 2Improve the models already present as they have not been op

Built With

• dash
• pandas
• plotly
• python
• scikit-learn