back to list

Project: Interactive Data Exploration in Healthcare Using a Local NL2Vis Approach

Description

Hospitals store vast amounts of patient and operational data within their internal databases. Accessing this information often requires specialized query knowledge, limiting its usability for medical staff and analysts. This project aims to develop an AI-powered natural language interface that allows users to ask questions in plain language and receive responses through interactive visualizations (e.g., plots, charts, or dashboards).

Building on concepts from the NL2VIS framework, this project focuses on developing a lightweight, locally deployable NLP framework that interprets user queries, maps them to relevant database fields, and constructs appropriate visualization outputs. The method will be designed with a strong emphasis on data privacy, interpretability, and efficiency, ensuring compatibility with hospital IT environments.

The project is conducted in collaboration with Ximius, which manages hospital data systems. The goal is to enhance the accessibility and usability of medical data for clinicians, administrators, and analysts through intuitive, explainable, and secure AI-driven visualization.

Details

Student

RR

Reuben Philip Roy

Supervisor

Fernando Paulovich