A mobile application that lets users smartly diagnose their symptoms by providing the most likely condition for the symptom, a summary of the condition (using web-scraping and a medical database) and the nearest location to receive treatment (through a geolocation API).
We were inspired to allow people to easily diagnose their medical symptoms and recieve the proper steps for treatment.
The users input their symptoms into an autocomplete search bar and receive their most likely diagnosis and severity of the condition through a machine learning API. Next, a web-scraper scrapes a medical database to retrieve the most relevant information about the condition, and displays it for the user. Finally, an API filters for the nearest locations that can provide the necessary treatment and provides the address.
We used Vue for the front-end of the app, Django for the back-end, and Selenium for the web-scraping portion. The database we used was malacards.org. Our APIs consisted of the APIMedic and the HERE-geocoder API.
Portability was an issue we faced because we were trying to coordinate code on different machines with different dependencies, especially when trying to make the webscraper as robust as possible. Accomplishments that we're proud of
We are proud to have programmed a fully-functional full-stack application in just twelve hours time.
We learned about developing a full-stack app, web-scraping in python, and general research practices to use when dealing with an API. We also got better at reading and understanding documentation.
We plan to use machine learning to optimize our diagnosis, and to find a more robust API for providing diagnoses and symptoms.
apimedic
django
django-rest-framework
nativescript
python
selenium
vue
here-geocoder
:)