Dev.to AI 🤖 Ai 👁 0 📖 2 min read

AI Wellness Coach: Exploring Camera-Based Wellness with Presage SmartSpectra

This is a submission for the Presage SmartSpectra SDK Challenge. What I Built I built AI Wellness Coach, a wellness application designed to bring different aspects of personal well-being together in one place. For thi

This is a submission for the Presage SmartSpectra SDK Challenge.

What I Built

I built AI Wellness Coach, a wellness application designed to bring different aspects of personal well-being together in one place.

For this project, I started integrating Presage SmartSpectra to explore how camera-based physiological signals could contribute to a more context-aware wellness experience.

The application combines:

Angular 19 for the frontend.

Python Flask for the main backend API.

Node.js as a supporting service for the SmartSpectra SDK integration.

MongoDB Atlas for application data and wellness history.

The goal is to explore how signals such as pulse rate, breathing rate, and HRV could complement existing wellness features.

Demo

I don't currently have a separate hosted demo URL. The project source code and setup details are available in the GitHub repository linked below.

The camera preview and camera-placement guidance are implemented. I am still validating the live physiological readings, so I am not presenting unverified measurements as working results.

Code

GitHub Repository:
https://github.com/pratikdevelop/AI-Wellness-Coach

The repository contains the Angular application, Flask backend, and supporting SmartSpectra integration service.

How It Responds to Human Signals

The SmartSpectra integration is intended to explore camera-based physiological signals, including:

Pulse rate: exploring how pulse information could contribute to wellness awareness.

Breathing rate: exploring breathing patterns as another wellness signal.

HRV: exploring heart-rate variability as additional physiological context.

The current implementation includes camera access and placement guidance. Live signal output is still being validated, and I plan to improve the measurement flow and display as the integration progresses.

Prize Categories

Best Thoughtful Real-World Use

I chose this direction because I believe physiological signals can be useful when presented responsibly as part of a broader wellness experience, rather than as isolated numbers.

Consent and Data Handling

Camera access is initiated by the user through the browser's camera permission flow.

The application is intended for personal wellness awareness, not medical diagnosis or treatment. I do not intend to use it for covert monitoring or make medical decisions based on its readings.

The exact handling of camera frames, SDK telemetry, measurement retention, and deletion needs further verification before I describe the integration as production-ready. I will make those details clear as the implementation is finalized.

Agent Session (Optional)

I used AI assistance during development to explore the integration architecture, troubleshoot the frontend and backend communication, and improve the implementation workflow.

📰 Read the original article on Dev.to AI

Originally published by Dev.to AI. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.