Face Pace

Inspiration

FacePace was inspired by the desire to make comprehensive health tracking accessible to everyone with a smartphone, eliminating the need for expensive hardware or time-consuming lab tests.

What it does

FacePace is a non-invasive way to track biomarkers using your phone camera. Users take a short video and a photo, and within moments, they receive a comprehensive health summary. This includes information about their pace of aging, cardiovascular health, brain health, sleep quality, and skin condition. The app provides insights into functional age, heart rate variability, cognitive engagement, and more, all from a simple interaction with your phone.

How we built it

Frontend: Next.js web app deployed on Vercel
Backend: Python, hosted on Heroku
Database: Supabase
AI Models: Mistral AI, Pixtral-8b model for advanced image and video analysis
Additional technologies: Docker, open-source Python packages
Video and image processing for biomarker extraction

Challenges we ran into

Designing an engaging user experience to distract users during the lengthy video capture required for analysis
Optimizing processing speed and managing multiple calls to the Mistral API efficiently
Ensuring consistency in outputs across different runs with the same data and for the same person with different images

Accomplishments that we're proud of

Developing a user-friendly interface for complex health data presentation
Integrating advanced AI models for accurate biomarker analysis
Creating a shareable and social health tracking experience
Achieving comparable or superior results to traditional non-LLM approaches in functional age determination

What we learned

State-of-the-art AI models can match or exceed traditional non-LLM approaches in determining functional age from images
Establishing user trust in AI-generated health insights is challenging and requires more than just data presentation
The importance of UX design in health tech applications, especially for data collection processes
Balancing accuracy with user experience is crucial in health-tracking applications

What's next for Face Pace

Significantly expand the range of biomarkers analyzed
Develop a native mobile app to improve image/video quality and enable integration with Apple Health data
Incorporate trusted inputs from other health data sources to enhance biomarker accuracy
Validate results against ground truth data from actual blood sample biohealth data
Explore partnerships with healthcare providers for clinical validation and application
Implement features for tracking health trends over time, providing users with long-term insights into their well-being