2026-05-13: Awarded Grant for AI-Driven Accessible Telehealth Support for Unhoused Populations

 

I am pleased to share that I have been awarded an Internal Research and Development Grant from the Office of Enterprise Research and Innovation at Old Dominion University in the amount of $50,000, beginning July 1, 2026. The project, titled “Trust-Centered AI Avatars for Accessible Telehealth Support Among Unhoused Populations,” is co-led with Drs. Ginger Watson and Tina Gustin.

This project proposes to improve healthcare access for vulnerable and underserved populations by leveraging, enhancing, and evaluating an NSF-funded telehealth kiosk, TeleHSupport, to improve usability through an audio interface and AI-powered interactive avatar. TeleHSupport is a secure, self-service kiosk enabling individuals with limited access to traditional healthcare to receive reliable AI-generated health information and guidance. It delivers safe, evidence-based guidance on chronic disease through text-based conversational AI, with clinically validated content curated by advanced practice nursing faculty. Individuals select symptoms through step-by-step prompts, which are processed using a retrieval-augmented generation approach to assess symptom severity and likely conditions. The kiosk provides consistent guidance while prioritizing patient safety. Ongoing deployment and focus group studies have identified low general and health literacy, and trust as barriers that limit adoption and effective use.To address these challenges, the proposed work integrates voice recognition with a human-AI avatar agent to enable an agentic AI framework, support multimodal, conversational interaction, reduce reliance on text, and foster trust, engagement, and comprehension. Leveraging generative AI, large language models (LLMs), Retrieval-Augmented Generation (RAG), human-computer interaction (HCI), and modeling and simulation, the project will design, implement, and evaluate avatar-mediated interactions tailored to the needs of unhoused populations.

The work builds directly on my prior NSF investments and represents a critical next step toward scalable, trustworthy AI-enabled health support systems deployable beyond Hampton Roads. The proposed study will provide usability and adoption data critical for larger-scale deployment within and beyond Hampton Roads.

Faryaneh Poursardar (@Faryane)

*Picture: Patient signed the consent form.