What I Built

Dr. T — Presage Sense is a camera-powered wellness experience integrated into Dr. T's Health Alert Mini platform.

It uses the Presage SmartSpectra SDK to transform camera input into physiological signals, including pulse rate, breathing rate and heart rate variability (HRV).

Instead of presenting measurements as isolated numbers, Presage Sense creates an interactive wellness experience:

  • A consent-first camera viewfinder with real-time optical signal visualization.
  • Live pulse, breathing-rate and HRV indicators.
  • Signal-quality feedback and camera-positioning guidance.
  • An animated 4-4-4 box-breathing pacer.
  • A subjective mood and perceived-stress check-in.
  • Gemini-powered educational reflections combining available measurements with user-reported feelings.
  • An optional session archive with individual record deletion and JSON export.

Presage Sense is a dedicated experience within Health Alert Mini, not a separate application. It connects wellness measurements with Dr. T's existing health-navigation experience while preserving the platform's other modules.

Demo

  • Live application:

https://ais-pre-4s4jvpipr3mh3mz6x2hpfp-393352619239.asia-southeast1.run.app#healthalert

The demonstration will show:

  1. Opening Presage Sense inside Health Alert Mini.
  2. Reading the camera consent information and granting permission.
  3. Starting a live SmartSpectra measurement.
  4. Viewing the pulse, breathing-rate, HRV and signal-quality indicators as supported by the SDK.
  5. Using the animated 4-4-4 breathing pacer.
  6. Completing the mood and perceived-stress check-in.
  7. Viewing the Gemini educational reflection.
  8. Importing a measured resting pulse into Health Alert Mini.
  9. Reviewing and deleting a saved session.

Code

  • Public repository:

https://github.com/Zenieverse/Dr.-T

The implementation and submission documentation are maintained in the existing Dr. T repository.

The project uses the existing Dr. T application architecture and integrates Presage Sense into Health Alert Mini.

How It Responds to Human Signals

Presage Sense integrates the native Presage SmartSpectra Node SDK (@smartspectra/node-sdk v3.4.0).

The implementation uses native FFI integration with the SmartSpectra C ABI library on Linux x64 and custom RGB video-frame input.

The application requests the supported physiological metrics and processes the returned metric payloads.

  • Pulse rate

SmartSpectra-derived pulse measurements are presented in BPM with rhythm-confidence and signal-quality feedback. The interface provides camera-positioning and lighting guidance when signal quality is insufficient.

  • Breathing rate

Breathing-rate measurements are displayed in breaths per minute. The experience uses these measurements to help users observe their breathing patterns.

  • Heart rate variability

When supported and sufficient measurement data is available, HRV is presented as RMSSD in milliseconds. It is shown as an informational physiological measurement, not a diagnosis or definitive stress assessment.

  • Real-time response

The interface combines the incoming optical signal visualization, measurement indicators and signal-quality feedback. Users can adjust positioning, improve lighting, pause or stop the session.

The 4-4-4 breathing pacer provides a guided breathing experience alongside the measurements.

  • Integration architecture

The backend initializes the native SDK, ingests camera-frame buffers, requests supported metrics and decodes the returned protobuf payloads. The diagnostic endpoint /api/presage/status reports engine state, requested metrics and API-key authorization status.

Gemini 3.8 Flash is used separately to generate educational reflections from available physiological measurements and the user's voluntary mood and perceived-stress check-in. Gemini does not replace SmartSpectra or generate physiological measurements.

Categories

We are entering both categories.

Most Creative or Delightful Use

Presage Sense explores a different way of experiencing human sensing: turning a camera-based physiological measurement into an interactive wellness check-in.

The experience combines optical signal visualization, live physiological indicators, a breathing pacer and personal reflection in one coherent interface.

Rather than making users interpret technical measurements alone, it makes the experience understandable and engaging through visual feedback and guided interaction.

Best Thoughtful Real-World Use

Presage Sense addresses the need for accessible, understandable wellness feedback.

It is designed for people who want to observe their physiological signals, reflect on how they feel and explore simple breathing exercises.

Its integration into Health Alert Mini makes the experience part of a broader health-navigation platform rather than an isolated sensing demonstration.

The design prioritizes informed consent, accessible instructions, signal-quality feedback, user control and privacy.

The feature is intended for general wellness and education, not medical diagnosis, treatment or clinical decision-making.

Consent and Data Handling

Camera access is gated behind an explicit informed-consent workflow. The user must grant permission before video tracks are initialized.

The interface provides a visible camera-active state and controls to pause, stop or exit the measurement experience. Camera tracks are disposed of when the session stops, pauses or exits.

The application does not require users to record or upload video as a default wellness-history feature.

Session records are stored separately in the presage_sense_sessions collection with user ownership restrictions. Users can delete individual records and export their session data as JSON.

Firebase security rules enforce owner-based access. API credentials remain server-side and are not exposed in frontend code.

Gemini 3.8 Flash is an optional AI service for educational reflections. Its use is distinct from the SmartSpectra measurement engine. Users should be informed when their session information is sent for AI processing.

The experience does not provide diagnoses, prescribe treatment, make emergency decisions or perform covert monitoring.

Thanks for playing with my Solution!