Careity | Devpost

Patient Onboarding
Data Intake
Patient Profile
Doctor Dashboard
Wellness Check Portal
Patient Detail View

Inspiration

Careity was inspired and informed by research insights on marginalized communities and their sense of security in adopting technologies, particularly AI. As a result, our product is centred on transparent systems that bridge the clinician-patient communication gap in a systematic and simplistic manner in order to mitigate the avoidable emergency hospitalization rate.

What it does

Careity uses a virtual intake assistant at its core, delivering a ready report to healthcare providers after a patient completes a non-invasive wellness check for clinical review. When a patient initially joins the app, they provide an array of health data (e.g., age, weight, height, hereditary risk factors, chronic illnesses, etc.). From there, the assistant will return a scheduled frequency of check-in notifications based on their health severity. At these intervals, users will be prompted to take a 'Wellness Check' through our app. This is a quick, friendly conversation with a virtual assistant where a patient provides updates on their physical or mental health, which in turn will be relayed to healthcare providers if necessary. This communication system, determined by severity of risk and missed appointment triggers, enables early risk flagging that “catches” problems before they develop to the point of hospitalization.

How we built it

Initially, we spent time understanding the problem, and brainstorming unique solutions to implement. Once we found an idea we liked, we used Figma to design and structure it, fleshing out all aspects of the solution. We then got to coding our prototype, and building our pitch deck. Finally, we polished our deliverables, practiced, and got ready to share out idea with the Spark team.

Challenges we ran into

In building our mobile application, we struggled a bit with implement our AI-powered virtual assistant. We had to learn how to use a third party API (openAI), and write structured, high quality prompts to tailor the LLM responses to fit our use case.

One major design/UX our team ran into while building was contrasting a colour theme across the clinical and user ends of the platforms. Surprisingly, this took us a considerable amount of time to sort, since clinical colour schemes tend to be more stark, yet we wanted the user to feel invited.

Accomplishments that we're proud of

We are proud of our unique and viable solution to the problem. We are proud of being able to build it out considerable in a short period of time, pushing past obstacles or tricky challenges that came up. We are proud of our perseverance and hard work!

Another milestone for our team was finishing ideating and prototyping early enough that we were able to revisit our design to “perfect” some pieces we were hoping to develop further (e.g. successfully integrating the chatbot into our technical prototype, the glass weather icons, etc.).

What we learned

This competition was an exciting challenge for us, as we were looking forward to sharpening our technical skills and collaborating together as friends. It was our first time building an AI-powered conversational tool, which came with a lot of mini lessons! We also learned about several new Figma features, and became all the more familiarized with the design platform. Most of all, through the challenges that came with implementing our chatbot and design consistency concerns, we practiced maintaining an eager attitude that we look forward to bringing to other hackathons!

What's next for Careity

Going forwards with Careity, we look forward to seeing how younger or middle-aged populations, such as parents, might use this kind of platform. For instance, we look forward to exploring how parents might be able to carry “multi-profiles” for their children. For the future, we would like to consider how connectivity issues might be incorporated into local updates (or not, alternatively). We also would like to incorporate more ideas on community care and Indigenous populations in order to increase the effectiveness and applicability of this app for residents in remote areas such as Clearwater Ridge.