PatientComm

App Logo

Inspiration

We were inspired by the courageous actions of healthcare workers during the Covid-19 pandemic, we decided that we wanted to do something to help them. We learned that communication errors between hospitals often resulted in an unfathomable amount of deaths. We decided that this issue could be solved in a way that also improves doctor-patient care by analyzing patterns.

What it does

Our application serves to provide medical information in laymen terms through web crawlers and distinct databases, connect patients and doctors across the world using real-time anonymous health care databases, and facilitates the exchange of past medical information for consulting or transition purposes.

How we built it

Our application utilizes Swift for the front end of the application, Python for the data handling and Machine Learning Models and Firebase API’s for data storage. We employ modern levels of encryption layered on top of our custom Blockchain framework in order to adhere to HIPPA guidelines.

Challenges we ran into

A primary issue of ours was employing our data searching API’s since we often received information that was full of jargon and replicated the problem we intended to solve. To combat this, we employed a supervised learning algorithm that would take in an index of the simplicity of data. After labeling the text, the ML algorithm picked up on more useful data and treatments that would be displayed to the patients. Another problem we faced was determining the number of groups for the K-Means Clustering algorithm. We employed an elbow test and used basic calculus to determine the optimal number of groups and set up an early stopping point so that patients would still have multiple choices.

Accomplishments that we're proud of

We were able to successfully identify a problem, model a solution, create a prototype app, as well as create an effective business model in the span of 3 days, and we had fun while doing it.