2.2.0 - ci-build

VoiceaBiomarkerforAI - Local Development build (v2.2.0) built by the FHIR (HL7® FHIR® Standard) Build Tools. See the Directory of published versions

Index

Official URL: https://kind-lab.github.io/vbai-fhir/ImplementationGuide/kindlab.fhir.vbai Version: 2.2.0
Draft as of 2024-05-29 Computable Name: VoiceaBiomarkerforAI

Introduction

​The Voice as a Biomarker for AI Health (VBAI) profiles are based on FHIR R4 and derived from US Core STU5 where possible. The profiles capture the data collected in the VBAI study. More information can be found at IntroToFhir.

What is Voice as a Biomarker for AI?

​The National Institue of Health (NIH) started the Bridge2AI initiative to fund the creation of foundational datasets in the medical field. Voice as a Biomarker for AI (VBAI) is one of five foundational datasets being constructed currently. The VBAI project will capture multimodal data from vocal recordings, genomic screens, waveform profiles, and electronic health records. The goal of the VBAI project is to create a dataset that can be used to create applications to diagnose conditions based on voice recordings. The researchers creating the VBAI dataset are multidisciplinary and across institutions in Canada, USA, and Mexico. More information on the VBAI project can be found at the main project website: voicecollab.ai.

Intended Audience

The VBAI Profiles are meant for use in research and development. The audience then breaks into two categories:

  1. Researchers: The final VBAI dataset can be used to test and explore novel FHIR applications related to voice as a biomarker
  2. Developers: The final VBAI dataset can be used to test the robustness of applications to real-world data.

Walkthrough of Implementation Guide