Inspiration
MyBetes was born out of a desire to simplify meal management for diabetics. Leif, one of our team members, is a type 1 diabetic and found it challenging to manage his blood sugar levels with just carb counting. Traditional methods didn’t always provide enough clarity or flexibility, especially when trying to predict how specific foods might affect his glucose levels. This gap in existing solutions led to the idea of MyBetes: a tool that offers personalized meal advice based on glycemic index, helping diabetics make better, more informed choices without adding unnecessary complexity to their daily routines.
What it does
MyBetes uses computer vision to analyze photos of meals and estimate their glycemic index (GI), which is a key factor in managing blood sugar. Once the meal’s GI is classified as low, moderate, or high, the app provides tailored advice for meal adjustments. For higher GI meals, the suggestions are more proactive, while lower GI meals get minor tweaks. Additionally, MyBetes includes a built-in assistant, powered by ChatGPT, that helps users with further insight and specific queries. The goal is to provide practical, diabetes-specific guidance that fits seamlessly into daily life without being overwhelming.
Challenges
One of the major challenges we faced during the hackathon was a third-party API we initially relied on going down unexpectedly. This setback forced us to quickly pivot to a new API, but it didn’t offer all the same features. To make up for the missing functionality, we developed some proprietary algorithms to supplement the new API, ensuring the app still delivered useful and accurate meal analysis. This added complexity, but ultimately, it pushed us to innovate and find creative solutions on the fly.
Accomplishments
We’re incredibly proud of how MyBetes came together in just 36 hours. The team managed to create a fully functional app with a smooth, responsive UI that integrates AI and computer vision in a way that’s easy for diabetics to use. Most importantly, we created an app that can make a real difference in the daily lives of people managing diabetes.
What we learned
Throughout the hackathon, we learned invaluable lessons about machine learning, web development, and working as a team under pressure. We also gained deeper insight into the technical challenges of integrating computer vision with backend systems and the importance of having a flexible approach when things don’t go as planned.
What's next for MyBetes
Looking ahead, we are considering continuing development and are especially excited about the possibility of shifting to a native iOS app. We believe a mobile version will make MyBetes even more accessible and user-friendly, and we’re eager to expand the app’s features to make it a more comprehensive tool for diabetes management. Our goal is to continue refining the app, adding personalization, and helping users live healthier, more informed lives.
Log in or sign up for Devpost to join the conversation.