CareCompass

Inspiration

People with medical disabilities, specifically those that are under recognized are not able to achieve the help that they need, this is because of barriers such as a lack of doctors that are available to aid them and a lack of information regarding their disease. What inspired us to create this tool was to help these people with the medical struggles they face because of their lack of recognition.

What it does

Our website is comprised of multiple components - The main component being a voice assistant for those with visual impairments who are unable to navigate the website without outside help, a medical decoder in which you can upload medical records and it'll be able to convert it to everyday language, aiding understanding about a user's current health, a specialized search engine for finding specialists and foundations dedicated to helping those with this specific disability, a personalized care plan for those who want to maintain their health at home and also helps schedule appointments and tracks symptoms, and finally, an AI care assistant that looks at a user's current medical situation and is able to make informed responses based on that.

How we built it

We built it by first drafting a detailed concept of what we're aiming to achieve including who we wanted to help, the judging criteria, and other concepts regarding how to help those with these diseases. Then we ran it through AI and did heavy research on the best solutions and databases for aid in order to further refine the concept and perfect it, then we ran it through GPT Sol 6.1 Codex to create a prototype of the website. After this, we made a log in order to detail beta testing experiences and the mistakes the website has and then used that log to create another prototype and then repeated that process until there was nothing to make a log about.

Challenges we ran into

We specifically ran into problems with the voice assistant, this includes it picking up noise from itself, background noise, and not being able to identify commands without it being verbatim. We also ran into problems with the voice assistant not being able to detail the website's contents enough to aid those with visual impairments.

Accomplishments that we're proud of

The agent that is integrated into the medical decoder is successfully able to interpret a various amount of conditions provided to it and outputs a proper simplification of it to those without knowledge of medical jargon, then additionally adds follow ups that detail what steps to take next in regards of the user's health. Another accomplishment that we're proud of is that when entering a health condition into the "Find a specialist" tab, it is able to successfully track down databases that have boatloads of medical information on the condition, and it even gives Google Maps links to medical specialists in the area of the ZIP code that is entered which further enhances user experience.

What we learned

We learned how to successfully integrate voice detection systems into websites and integrate/use AI to give personalized and detailed feedback when given user information.

What's next for CareCompass

Expanding the database and care assistant function.

Built With

• codex
• css
• html5
• javascript