Inspiration
Managing chronic conditions such as kidney disease, diabetes, and hypertension is overwhelming for both patients and caregivers. Diet and lifestyle are critical to long-term health, yet most people leave the doctor’s office with generic instructions that are hard to follow day-to-day. We wanted to create a digital companion that translates complex medical guidance into simple, personalized, and supportive suggestions that can be used at home.
What it does
PHA (Personal Health Assistant), formerly called NephroNom, is an AI-powered web app that provides patients with tailored recipes, diet suggestions, and lifestyle tips. By inputting basic health information or dietary restrictions, users receive recommendations that are safe, easy to understand, and relevant to their condition. The goal is to give patients more confidence and control over their daily health decisions.
How we built it
We built the app using HTML and CSS for a lightweight front end, with GPT APIs powering recipe generation and health suggestions. Collaboration was managed through GitHub, where we iterated quickly, merged contributions, and tested features in real time. The focus was on simplicity—delivering a working proof-of-concept that patients could realistically use.
Challenges we ran into
One major challenge was ensuring that AI outputs were safe, consistent, and medically appropriate. Large language models are stochastic by nature, so we had to carefully design prompts and workflows to reduce errors. Time constraints also meant making tough choices between adding advanced features and delivering a functional MVP within the hackathon window.
Accomplishments that we're proud of
- Building our first fully functioning app as a team.
- Translating a complex healthcare problem into a working prototype in under 36 hours.
- Receiving positive feedback from mentors and peers about the real-world potential of PHA.
What we learned
We learned how to collaborate effectively under tight deadlines, how to prototype an AI-driven application quickly, and how to transform medical pain points into technical solutions. We also came away with a deeper appreciation for the need to validate healthcare apps with professional input.
What's next for NephroNom / PHA - Personal Health Assistant
We see PHA as the beginning of a broader vision. Next steps include:
- Partnering with dietitians and healthcare professionals to validate outputs.
- Expanding beyond kidney disease into conditions like diabetes and hypertension.
- Building a mobile app version for better accessibility.
- Exploring wearable integrations to provide real-time feedback.
We’re excited to see where this project can go and how it might one day empower patients and caregivers worldwide.
Log in or sign up for Devpost to join the conversation.