Team Members - Telvin Zhong, Ethan Domingo, and Mareesh Kumar Issar
For this hackathon, we wanted to create something that would increase accessibility to convenient COVID-19 screening. Most existing screening services require stable internet connections that are not always available to the socioeconomically disadvantaged, despite being the group most at risk for COVID-19. Additionally, websites optimized for desktop usage with too many options may be more difficult to navigate to for senior citizens. Our goal was to create a streamlined approach for screening COVID-19 symptoms with fewer practical barriers and technical complications.
What it does
Our team created hotline services that can be either called or texted to help users screen themselves or others for COVID-19 based on factors including but not limited to symptoms, recent travel habits, and pre-existing health conditions.
The hotline will make one of four recommendations along with options for additional information on staying safe.
Based on their responses, those at minimum risk will be asked to continue observing social distancing.
Those at low to medium risk will be recommended to quarantine themselves at home for at least 14 days.
Those with medium risk will be warned to self-isolate and to be prepared to seek medical assistance should symptoms worsen.
Those at medium to high risk will be advised to contact their healthcare providers and be tested for COVID-19. We've also created a chatbot that can detect key words (by voice or by text) and help guide users towards relevant resources, including recommendations on fun actvities to perform indoors and good health practices to maintain. The chatbot can also directly link users to both the text and phone hotlines for our screening service.
To access the SMS screening service, text COVID to + 1 669 800 5670.
To access the IVR screening service, please call +1 310 494 2720.
To access chatbot please text COVID19BOT to +1 3233100369.
How we built it
We used Twilio APIs to secure phone numbers that could respond to texts and calls. The programs themselves were built with Python, Flask, ngrok, and JSON, allowing them to respond to user inputs based on their responses.
Challenges we ran into
Professional screening service algorithms used to assess risk based on questions are not public. We opted to use trial and error approaches to get our programs to consistently output the same results as professional services. Team member time zone differences also added a degree of difficulty in collaboration.
Accomplishments that we're proud of
Having a complete ecosystem of products for COVID-19 screening.
Ethan and Telvin would like to announce that this was their first hackathon and that they hope you find the product useful! Telvin has also never never coded until two months prior to this hackathon.
What we learned
Collaboration. Software design. Webhooks and APIs
What's next for QuaranTwilio
Acquiring feedback on the results of our service from healthcare professionals. Launching our service as a full fledged product in multiple regions across the world.