Fhirband

The ultimate wearable for first responders, with AI-driven vitals & insights to optimize safety & performance. Its powered by our self-built platform—with seamless integration for new medical devices.

Comment

Inspiration

For Angelenos, fires are far from unexpected. Small fires happen, but it was always "under control." Along the Eaton Canyon trails, my friends and I would take night hikes all through senior year, reflecting on old memories and speculating about our futures in college. The Palisades and Eaton fires changed everything.

The Eaton fire not only set ablaze the mountains and memories I grew up on but devastated the community as a whole: friends without houses, scrambling business-owners, and, mostly unseen, neglected first responders. Growing up next to a fire station, I saw firefighters -- the first line of defense -- during these brutal fires facing dangerous, extreme conditions without immediate health insights and support.

Enter the idea for Fhirband, a wearable device that integrates seamlessly with our self-built medical platform, allowing for real-time AI-driven insights and squad-level health analytics. During disasters, firefighters and emergency teams struggled with real-time health monitoring, often pushing their bodies to the limit without immediate feedback on their vitals. In conjunction with our overall mission to bridge testing and treatment for medical devices, Fhirband is a cornerstone to creating an open, interoperable platform to unify health diagnostics.

What it does

Fhirband is an AI-powered wearable designed for first responders, providing:

  • Real-time vitals tracking: Continuous monitoring of heart rate, body temperature, humidity, and air pressure.
  • Squad health insights: Aggregates biometrics to provide commanders with strategic squad-level health data.
  • Adaptive haptic feedback: Alerts users when vitals indicate danger, enabling quick response.
  • Seamless software integration: Built on our universal medical device platform, making it the go-to solution for any new IoMT wearables.

How we built it

Fhirband was developed using:

  • Adafruit ESP32-S3 Reverse TFT Feather (2 units) – Low-power microcontrollers for real-time data processing.
  • Raspberry Pi 4 Model B (4GB) – Central processing hub for AI-driven health insights.
  • Heart Rate Sensor Module (MAX30102) & Pulse Sensor Kit – Tracks oxygen saturation and cardiovascular health.
  • Adafruit BME688 Environmental Sensor & AMG8833 IR Thermal Camera – Monitors air quality, temperature, and heat stress.
  • TCA9548A I2C Multiplexer – Handles multiple sensors efficiently.
  • Adafruit DRV2605L Haptic Motor Controller & Vibrating Mini Motor Disc – Provides haptic alerts for real-time safety warnings.
  • Lithium Ion Polymer Batteries – Ensures long battery life for extended use in the field.
  • Google Cloud - Gemini AI – Processes real-time biometrics with multimodal AI for squad-level insights.

On the software side, we use our unified IoMT platform, Medibound, for diagnostics, integration, and telehealth to power Fhirband and any new medical devices. (To be clear, Medibound is not our project. Fhirband is our project and it's powered by Medibound. Details regarding Medibound are provided only to explain the platform Fhirband operates on.)

Backend Architecture & Data Standardization

  • No-Code Backend & Partner Portal
    • Drag-and-drop IoMT integration platform, allowing manufacturers to onboard devices in minutes.
    • Supports Bluetooth Low Energy (BLE) & Near Field Communication (NFC) protocols.
    • FHIR4-Compliant Data Standardization
    • All device data is automatically converted to FHIR4, ensuring seamless interoperability with Electronic Health Records (EHRs).
      -Google Cloud’s FHIR API, making it scalable & compliant with HIPAA & FDA regulations.

AI-Driven Diagnostics & Insights

  • "Medi" AI Health Assistant
    • Built using Gemini AI models, Medi analyzes real-time vitals, detects abnormalities, and provides personalized health recommendations.
  • Predictive Analytics & Automated Risk Scoring
    • Developed custom AI algorithms to identify early signs of fatigue using wearable sensor data

Frontend Development & User Experience

  • Common Health Record UI
    • Created a cross-platform dashboard aggregating health data from multiple devices into an intuitive visualization interface.
    • Built using FlutterFlow & React, ensuring real-time interactivity across web & mobile.
  • Secure Telehealth Integration
    • Developed FHIR-compliant APIs that allow doctors & first responders to remotely access patient vitals in real-time.

Security & Compliance

  • HIPAA-Compliant Encryption & Regulatory Safeguards

By building Fhirband’s cutting-edge wearable technology with our scalable AI-driven platform, we’ve created a seamless, interoperable IoMT ecosystem that bridges the gap between diagnostics, device integration, and clinical decision-making.

Challenges we faced

  • Data Synchronization: Merging multiple sensor streams in real time without lag.
  • Power Efficiency: Ensuring long battery life while maintaining constant monitoring.
  • AI Processing: Running AI models on-device without cloud dependency for immediate feedback.

Accomplishments that we're proud of

  • Developed a fully functional prototype with real-time AI-driven vitals monitoring.
  • Created an adaptive AI assistant to provide live health insights.
  • Achieved seamless interoperability with our broader medical device platform, making Medibound usable across any IoMT wearables.

What we learned

  • AI can operate efficiently for real-time health insights without cloud dependence.
  • Squad-level biometric monitoring has untapped potential in emergency response, military, and everyday consumer use.
  • First responders need intuitive, non-intrusive alerts, making haptic feedback critical.

What's next for Fhirband.

  • Enhancing AI diagnostics: Continuing to use Gemini AI to refine real-time health predictions.
  • Scaling to field tests: Partnering with fire departments and other units for real-world validation.

Built With

  • ai
  • ble
  • edge-computing
  • ehr
  • fhir
  • flutterflow
  • gemini
  • generativeai
  • hardware
  • healthcare
  • iomt
  • medtech
  • nfc
  • raspberrypi
  • react
  • sensors
  • software
  • spreadsheet
  • ux
  • wearable
Share this project:

Updates