Entera

Inspiration

Waking up with Crohn's disease is telling a story without change; every morning I eat eggs, every lunch I have chicken, and by dinner, if I dare to try something new, I must live with the risk of exacerbating pain for the next 4 days. Crohn's, as with many other gastrointestinal diseases, severely limits the type of food I can eat. Unfortunately, no illness comes with a guidebook, and the task of mapping dietary restrictions is unique for every individual. This has inspired my team to create Entera.

What it does

Entera helps users track and predict dietary triggers associated with Crohn’s disease and other gastrointestinal conditions. Symptoms can become noticeable between 6 and 72 hours after consumption, so it becomes nearly impossible to identify which specific food is provoking a reaction after several meals across multiple days. This problem can be analogized to a system of equations with dozens of variables, which we solve through our algorithim Moreover, the app collects dietary information across all users and pools data that can be used towards research to find cures for these diseases.

How we built it

We built a mobile app with a frontend for logging meals and symptoms, plus a Supabase database storing timestamps, severity, and per-food trigger probabilities. Users enter meals in natural language, and an LLM extracts base foods/spices into JSON. A scoring algorithm updates trigger probabilities and displays high/medium/low risk with “safe/caution/avoid” guidance.

Challenges we ran into

One of our biggest challenges was aligning everyone’s ideas. We each had different perspectives on how the app should work, and combining medical concepts, LLM behavior, and data logic wasn’t straightforward. It took discussion, compromise, and iteration to merge our ideas into a single, cohesive design that everyone agreed on.

Accomplishments that we're proud of

Creating a product that doesn't exist that has the potential to change the way people live with IBD and IBS.

What's next for Entera

We want to move beyond simply identifying triggers and focus on prevention. With data from thousands of users, we aim to uncover shared patterns and common triggers across different IBS, IBD, and Crohn’s subtypes. Over time, this growing dataset can be used to continuously train more advanced neural network models, improving accuracy and personalization for users. Our long term goal is to support earlier intervention, better symptom management, and meaningful contributions to research that move the field toward preventative care and, eventually, a cure.

Built With

• expo.io
• node.js
• openrouter/openai
• python
• reactnative
• sql
• supabase
• typescript