ML Doctor

• 

  The front end of our website where you can access both the diabetes detection page built with jupyter notebook and corona detection page.

Inspiration: We built this project because in our current situation, the coronavirus is becoming a a huge issue, and cases do not seem to be decreasing any time soon. The goal of our project is to decrease the test time for people all around the world by using multiple thousands of datasets to train our algorithm, ensuring that a false-positive result will not occur.

What it does: ML Doctor allows the user to enter in several symptoms that the suer has, and in turn gives the chance of having a certain disease, such as the coronavirus. Furthermore, other diseases can also be detected using this site, including Breast Cancer, and even Diabetes. This especially helps many people in low income areas since hospitals are crowded in those locations, and not many tests are administered in order to control the spread of the coronavirus. By using our website, people will be more willing to quarantine themselves if a positive result comes up, decreasing the spread of the coronavirus, even if it is just a small step.

How we built it: We built the website using HTML/CSS/JS. We build our Diabetes detection site using the Jupyter notebook and Python. We built out corona testing application using HTML/CSS/JS. We trained our algorithm for checking for the disease with tensorflow using several thousand data sets we acquired from both Kaggle and the Google cloud base.

Challenges we ran into: Some challenges we ran into along the way was figuring out how to train our data using tensorflow. This was our first time training a Machine Learning model, so it was not easy to find out how to train our model and incorporate it with our website without having to make significant changes to our design and functionality.

Accomplishments that we're proud of: We are proud of actually being able to build a machine learning tool that is functional and one that has a high chance of returning trustable results.

What we learned: We learned that building a machine learning model takes a long time. Training our model with several thousand data sets definitely took hours to accomplish, but at the end, the time put forth into this project was worth it.

What's next for ML Doctor: Our next objective is to expand the diseases ML Doctor can detect. By finding more datasets that we can train to make our machine learning model for ML Doctor more efficient, we can definitely impact a lot of low income areas that cannot afford testing, and even save many lives.

Built With

• css
• googledata
• html
• javascript
• jupyter
• kaggle
• python
• tensorflow