Travelling to a new country can be a daunting experience. One of the most stressful parts of travelling abroad is a medical issue. Moreover, for international students like us, who are moving to a new country for education, navigating healthcare is a difficult task. Every country has a different healthcare system and for an international or a non-citizen, making use of the health facilities can be an arduous experience. PhysiBot is built with the aim to make this process simpler.
What it does
PhysiBot is a one-of-kind chatbot that is built to be your health companion when you travel anywhere. PhysiBot is an assistive app developed with the aim to provide a preliminary objective diagnosis based on our state-of-the-art machine learning model and recommend the nearest healthcare facility using geomapping while travelling in a foreign country. Through its interactive and intuitive UI. PhysiBot provides a seamless healthcare experience while travelling.
How I built it
We ran multiple one-hot encoding and cleaning procedures to refine the raw data. Afterwards, we put the refined data set into the IBM Cloud Watson Studio. Using the Studio, we were able to choose and build the best model for our project and refine it using various pipelines, all with so much ease. IBM Cloud Services helped us phenomenally. In Watson's machine learning, we used the XGB classifier. We built the webpage in React.
Challenges I ran into
The first and foremost challenge was collaborating in a virtual environment. All of us had participated in hackathons before as part of different teams, however, this was our first hackathon which was completely virtual. We also had to learn how to use the IBM Watson Studio as part of the IBM Cloud package, but that was an easy process thanks to the option of learning with a project demo and interactive tutorial on the website.
Accomplishments that I'm proud of
We’re proud of the fact that we were able to collaborate well with each other in a virtual environment despite Hack the North being the first hackathon we are participating in together as a team. All three of us are freshmen at the University of Waterloo and haven't even met each other in person yet! We were able to communicate and work jointly to achieve our goal in the given time owing to services like Trello, Discord, Zoom etc. We are also pleased with the product that we’ve built. The fact that we have the potential to embellish the socio-cultural fabric of society with an algorithmic sapphire that we have crafted and consequently, improve people’s lives for the better is especially thrilling.
What I learned
This whole experience was eye-opening in terms of educating us about the problems that exist, waiting to be solved. Consequently, we also learnt how to harness the might of computer science and the technology available to us to solve some of these problems and enhance the current human condition. Moreover, we are also grateful that we were able to gain an insight into virtual coworking and collaboration that will undoubtedly help us as industry professionals after a few years. We also learned about the importance of machine learning models, using the IBM Watson API and its numerous applications which allow users to build extremely accurate ML models in a matter of hours through their intuitive and interactive UI/UX.
What's next for PhysiBot
As for the future, we plan on refining our model to further judge the urgency of the user’s medical condition using more parameters.
We also envision expanding our product by adding medical certificates and medical insurance documents inside the app using a secure portal. There exists no common framework where you can store all your medical certificates and insurance in one place. Our product will be used at borders, airports to conduct health checks and in hospitals while verifying insurance claims. The potential to scale our app into a product or startup is extremely high and that will also help us build an edge over other potential competitors in the market.
We first plan to expand and test our service for the South-Asian and Middle-East markets since all of us come from there and we understand the endemic markets. We plan on launching our beta with a two-dimensional business plan - firstly, a subscription model for students like us who study abroad and live there for a long period of time. Secondly, a one-time cost plan for families and people who are planning a trip and don’t go abroad periodically.