FocusTellisense AI: Making Human Signals More Visible and Support More Accessible
This is my submission for the Presage SmartSpectra SDK Challenge.
What I Built
FocusTellisense AI is a camera-based physiological monitoring application built to explore how technology can make human wellbeing more visible, understandable, and accessible.
People do not always recognise changes in their physical state or know when they might need to pause, rest, or seek assistance. This is especially relevant to older adults, people living alone, and individuals who may benefit from additional support.
I wanted to explore a simple but meaningful question:
What if an ordinary camera could help people become more aware of what their bodies are communicating?
Using the Presage SmartSpectra SDK, FocusTellisense AI is designed to work with camera-derived physiological signals, including:
- Pulse rate: Observing changes in heart rate.
- Breathing rate: Providing visibility into breathing patterns.
- Heart rate variability (HRV): Exploring another dimension of physiological activity.
The application brings these signals into a visual dashboard with individual monitoring panels and graphs.
Built with Electron, React, and TypeScript, it includes camera monitoring controls, a dedicated signal-monitoring interface, and a clearly labelled development simulation mode.
My longer-term vision is to evolve FocusTellisense AI into an adaptive human-support experience that can help people recognise changes in their wellbeing and connect with appropriate human assistance when needed.
Demo
Watch the demonstration:
The demonstration introduces the FocusTellisense AI interface, its monitoring workflow, signal dashboard, and development simulation mode.
The simulation mode is explicitly identified as development data. It is separate from genuine SmartSpectra measurements and is not presented as live physiological data.
Code
GitHub repository: FocusTellisense AI
The repository contains the application source code, Electron desktop architecture, React interface, TypeScript implementation, and project setup instructions.
The project separates the desktop application, renderer interface, and supporting SDK integration to make the application easier to maintain and extend.
How It Responds to Human Signals
The central technology behind FocusTellisense AI is the Presage SmartSpectra SDK.
The project explores how camera-based physiological measurements can become part of an accessible, user-facing experience without requiring dedicated wearable hardware.
The monitoring experience is organised around three signals:
1. Pulse rate
Pulse measurements can help users observe changes in their heart rate during a monitoring session.
2. Breathing rate
Breathing measurements provide visibility into changes in breathing patterns.
3. Heart rate variability
HRV adds another physiological measurement that can be presented alongside pulse and breathing information.
The application is designed to connect the SmartSpectra measurement workflow to the React dashboard, where available readings are displayed through individual signal panels and visual graphs.
Users can initiate and stop monitoring through the application controls. The interface also includes a development simulation mode, allowing the dashboard and its interactions to be explored when live measurements are unavailable.
An important distinction in the implementation is that simulated readings are clearly labelled. They are not treated as genuine Presage measurements.
The goal is not simply to collect physiological numbers. It is to explore how these signals can be presented in ways that help people understand their bodies and make more informed decisions about seeking support.
Prize Categories
Best Thoughtful Real-World Use
FocusTellisense AI is my entry for Best Thoughtful Real-World Use.
The inspiration came from a real-world challenge: people may not always recognise changes in their physical state, particularly when they live alone or have limited access to continuous support.
I wanted to explore whether camera-based physiological sensing could contribute to a more accessible way of understanding human wellbeing.
The project is guided by three principles:
Awareness: Make physiological signals more visible and easier to understand.
Accessibility: Explore monitoring through a familiar device, without requiring specialised wearable hardware.
Human support: Build toward an experience that encourages people to seek appropriate assistance when they may need it.
The broader vision is to make technology more responsive to human needs, rather than expecting people to adapt to complicated technology.
FocusTellisense AI is not a diagnostic tool. It does not claim to diagnose medical conditions, recommend treatment, make medical decisions, or automatically contact emergency services.
It is an exploration of how physiological signals can support personal awareness and more responsive human assistance.
Consent and Data Handling
Privacy, transparency, and user control are important to FocusTellisense AI.
The application uses a user-initiated camera monitoring workflow. Camera access requires permission, and users have controls to start and stop monitoring.
The application maintains a limited in-memory history for displaying signal graphs during a session. Its monitoring cleanup workflow is designed to stop camera media tracks, stop the SDK, and clear session readings when monitoring ends or the monitoring component is unmounted.
The application is not designed for covert or non-consensual monitoring.
Presage SmartSpectra is used for the intended physiological measurement workflow. Optional SDK telemetry is disabled in the application configuration where supported. Any additional network communication required by the SDK or its licensing services remains subject to the SDK's configuration and service requirements.
The project does not intentionally present development simulation data as real measurements.
FocusTellisense AI does not claim to provide medical diagnosis, treatment, or automatic emergency intervention. Physiological measurements are presented for awareness and exploration, not as a substitute for professional healthcare.
What possibilities do you see for camera-based physiological monitoring in everyday life?
I'd love to hear your thoughts on how technology like FocusTellisense AI could make human support more accessible, especially for people living alone or those who need additional assistance.
Top comments (0)