Symp-See

Inspiration

Our interests lie at the intersection of health and technology, so naturally the HealthTech track was perfect for us! Stefany had an idea to create an app that documents patient symptoms, while giving them sort of analysis/prediction of symptom. The goal of this project was to use object-oriented recognition that would allow someone to take a picture of some part of their body that is showing a physical symptom (ex: rash, cold sore, etc) and Symp-See would output a possible match, with a blurb on common treatments. This app is NOT meant to serve as a diagnosis tool. Additionally, an option to contact/make an appointment with a doctor would be available. To better improve communication between patient/doctor, (in the future) Symp-See would be able to document the symptoms of the user overtime by taking pictures and adding captions, so that the doctor would have a log of the user's symptoms.

Not all the capabilities were created, but a starting framework for Symp-See was built.

What it does

Currently, Symp-See will prompt a User to input a picture into the web application. This picture will be sent to the back-end, which is powered by Watson Studio Visual Recognition, to determine if it matches one of the six possible matches (Cold Sore, Poison Ivy, Vitiligo, Melasma, Impetigo, and Squamous Cell Carcinoma). The front-end and back-end are not connected, so there won't be an output analysis for the user. However, the front end would have returned a blurb on the match, symptom descriptions, and possible treatments.

How I built it

Front end: HTML/CSS/JavaScript

Back-end: Python, IBM Watson Studio: Visual Recognition

Challenges I ran into

(This is the first hackathon for most of the members) Back-end: At first we tried finding an image scraping API, however most of them were not available or were costly. Object recognition (cam): Tried to borrow a camera from the hardware, however there were no adaptors/no other options. Front-end: general bugs

Accomplishments that I'm proud of

Front end: great interface design Back-end: framework

What I learned

We all learned/ got more experience with programming and specifically working on a project in HealthTech.

What's next for Symp-See

In the future, we hope to work on the capabilities noted above. We noticed that the object recognition component of our project was missing in most, if not all, the apps we researched before starting the project. Most of the other apps, that surveyed users symptoms and medical history had little visual components, and most times, no way for a user to document details related to their symptoms.

We feel that Symp-See has the potential to help make meetings with patients and doctors mutually beneficial, by serving as an aid for communication.