Hearta.tech

• 

  Patient View of their Records

• 

  Doctor's Dashboard for reviewing high risk patients

• 

  Patient Record Creation

• 

  Easy to use login page for both doctors and patients

• 

  User friendly sign up page, *doctors need to be approved separately*

• 

  Our database, currently having about 300 entries

Inspiration

We were inspired the kind of work Epic has done, creating million of records and vast databases that are monumental in supporting the healthcare industry

What it does

We create patient records based on lab results of certain parameters are that crucial for a healthy heart. We then use our AI model to predict the risk of heart problems, and help prioritize heart-care for those whose lives may depend on it.

How we built it

We built a custom Keras Sequential neural network model trained over data sourced from Kaggle. We custom built the backend and frontend interfaces connecting the doctors, patient and our AI model.

Challenges we ran into

We had a hard time configuring nginx and Django, our backend API. After about 6 hours of trying to connect a Google Cloud VM to our backend and failing, we decided to no longer pursue that route.

Accomplishments that we're proud of

We made a complete backend and user friendly frontend which serves dynamic content, while struggling to configure a server in the cloud to host our website.

What we learned

Using Tensorflow and Keras, HDF5 files, Django and Jinja templates

What's next for Hearta-tech

Integrate real-time communication among doctors and patients, use the growing database for further training,

Built With

• css3
• django
• html5
• javascript
• keras
• numpy
• python
• scipy
• tensorflow