Publication Date

Spring 2026

Degree Type

Master's Project

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

First Advisor

Vuthea Chheang

Second Advisor

Amith Kamath Belman

Third Advisor

Mira Jane

Keywords

Virtual Reality (VR), Large Language Model (LLM), 3D Environment, Natural Language Interaction

Abstract

Traditional approaches to learning and simulation often struggle to convey the complexity of three dimensional structures. In anatomy education, methods such as textbooks, lectures, and cadaver based learning are used, but can make it difficult to understand the complex structure of organs within the human body. Virtual Reality (VR) offers a promising alternative by providing immersive and interactive 3D environments that allow users to explore and manipulate anatomical structures more intuitively. These environments can be further enhanced by integrating large language models (LLMs) as intelligent agents capable of guiding users through natural language interaction. This project presents a Unity-based VR system in which an LLM acts as an intelligent agent that can interpret user input and interact directly with the virtual environment. The system enables users to perform tasks such as loading patient specific liver models, highlighting anatomical regions, and retrieving and displaying patient specific information through conversational commands. By combining VR with an LLM that serves as an intelligent assistant, the system allows users to interact with a 3D environment in a new way that may be more intuitive and flexible compared to traditional interfaces. Together, these technologies create an interactive environment that enhances user engagement and accessibility in anatomy education. Beyond this specific application, the project demonstrates the broader potential of integrating AI-driven agents with immersive environments, enabling more natural and efficient interaction with 3D systems across many domains such as education, scientific visualization, gaming, etc.

Download

Share

COinS