A friend, after his experiences, initially inspired us to pursue this project. While fortunately recovering, he mentioned that he did not quite grasp the full gravity of his mental condition. We invested our competition resources into crafting a capable and unique application that was specifically tailored for this need. The name of our application, Mentis, is derived from the Latin word for "Mental."

What it does

This Progressive Web App has three main functions:

  1. Describe what depression is, and let people know that it can be treated
  2. Administers a test to determine if someone likely has a depression. It uses a neural network to determine if/how depressed someone is based on known cases from medical professionals.
  3. Provides a one-on-one chatting system for users to talk with either volunteer peers or medical professionals. This chat system is moderated with a neural network, so hurtful comments cannot be made during the chat which could set off an episode. The app also recommends 911 or the National Suicide Outline for additional counseling which only a professional may be able to provide.
    ## How we built it We built it using Progressive Web App (PWA) technologies so that users can just add it to the home screen after navigation to the site. From there, the app then feels like a normal app installed from the App Store or Play Store. We used Node.JS to do the processing and REST APIs. We used the Riot.JS UI library to make custom components such as the bottom and top bar, and the message boxes. ## Challenges we ran into A primary example of a challenge that we encountered was the live chat functionality between two users in a lobby. This complication was resolved using a library called SocketIO, which conclusively handles primitive socket connections between servers and clients. Furthermore, to prevent the inevitable abuse of users within the chat lobbies, we created a custom moderation system that utilizes neural networks (similar to the survey) to automatically flag and deny any malicious comments. Finally, developing a proper UI/UX implementation that was functional in both PWA and browser formats proved itself as a worthy challenge due to the variety of devices. ## Accomplishments that we're proud of This was our first time working with sockets and a real time communication system. We are proud of making a communication system which is fast and scalable and can help save lives. ## What we learned We learned a lot about the component model of Riot.JS and other UI systems, and we learned how beneficial that these reusable modules can be for code organization and application speed. We also learned about socket communications and implementing rapid chat systems. We also learned a lot about neural networks and how they can be so beneficial in rapidly analyzing different different data sets and making predictions (in this case, taking in the answers to the questions and analyzing them to determine if someone is depressed). We also learned a lot about the Google OAuth API, Brain.JS neural networks, and ExpressJS. ## What's next for Mentis We would like to market this application to bring it to a wider audience such as high schoolers, populations who are susceptible to depression. We believe that this application can provide a new type of community-based therapy to remind those with depression that they are not alone and have a community to aid them.

Built With

Share this project: