Generative AI for Enhanced Healthcare Information Access
Target Students
- Computer Science (Bachelor/Master)
- Psychology (Bachelor/Master)
Requirements
- Computer Science:
- Python/JavaScript programming
- Basic AI/machine learning knowledge
- Psychology:
- Experimental design & evaluation
- UX research methods
Description
Modern healthcare systems require efficient access to medical information for both professionals and patients. While existing search engines like PubMed provide basic functionality, our prototype system (Wispermed Search Engine) has demonstrated superior performance in clinical studies with medical practitioners.
This project focuses on enhancing medical information retrieval through generative AI techniques. The core challenge lies in processing complex medical queries and providing traceable, aspect-specific summaries from scientific literature. Unlike conventional approaches that match keywords, our system will employ Large Language Models (LLMs) to understand contextual relationships and extract medical facets such as treatment protocols, patient demographics, and outcome measures.
A key innovation is the development of interactive verification mechanisms, allowing users to validate AI-generated insights against original source materials. This transparency feature aims to build trust in AI-assisted medical decision making. The project builds on existing infrastructure from the WisPerMed research group (RTG), including a working search engine and annotated medical corpora.
Send an email to Sameh Frihat with the following information: registration number, first and last name, course of study and semester. Registration is open from March 1st, 2025 (first come, first served).