Skip to main content

Vitals AI

Overview​

The Vitals AI SDK is a powerful tool that allows you to measure the vital signs of a user by analyzing a video of their face. The front-end SDK interacts with a websocket to support live data streaming to our back-end.

Prerequisites​

To use the Vitals AI SDK, you must have an account with Helfie.ai and request an API key. Please contact our support team to get started.

Websocket URLs​

An API key is required to use the following socket URLs.

  • Development Socket URL: wss://vm-development.xyz/vp/bgr_signal_socket
  • Production Socket URL: wss://vm-production.xyz/vp/bgr_signal_socket

Available SDKs​

  • iOS SDK
  • Android SDK
  • Web SDK
  • React Native SDK
  • Flutter SDK

How our SDK works​

  • Integration Process Overview:
    • FrontEnd initiates interaction
      • Sends integration request to SDK
    • SDK Operations:
      • Initializes itself
      • Authenticates via Auth
      • Opens a websocket connection with Helfie Cloud Engine
      • Starts the health scan process
    • UI/UX Checks Loop:
      • Checks if the face is detected
      • Checks if the face remains within the frame
      • Verifies the face is in the correct position
      • Ensures the face has the correct orientation
      • Confirms the face is under appropriate lighting conditions
      • Returns UI/UX check results to FrontEnd
    • Data Handling:
      • SDK sends BGR data to Helfie Cloud Engine
      • Helfie Cloud Engine processes and returns JSON results including metrics like Heart Rate, Respiration and others
      • SDK receives and relays the analysis results to FrontEnd for display
      • Helfie Cloud Engine also sends back status/error codes as necessary

URL construction​

1. Image processing websocket​

URL: ${BASE_URL}/socket?authToken=${authTokenValue}&fps=${fpsValue}&age=${age}&height=${height}&weight=${weight}&deviceInfo=${deviceInfoValue}

  • authTokenValue: Mandatory Auth Token.
  • fpsValue: [Optional] (int) Frames per Second target e.g. 30 (30fps);
  • deviceInfoValue: [Optional] (string) device name;
  • age: [Optional] (int) User age;
  • height: [Optional] (int) User height in cm;
  • weight: [Optional] (int) User weight in kg;

Frontend should open image processing websocket and send image files containing frames of user video stream. Backend sends back text messages with calculated parameters.