Abstract
With healthcare systems under growing pressure from rising patient volumes and shrinking consultation windows, improving how patients communicate with physicians has become essential to delivering quality care. Yet patients routinely arrive at appointments unable to clearly describe their symptoms, recall their medical history, or articulate concerns, contributing to miscommunication, diagnostic inefficiency, and pre-visit anxiety. This study introduces PreVisit AI, a conversational system designed to address this gap through structured, knowledge-based patient preparation. The system is built on a Retrieval-Augmented Generation (RAG) architecture combining HuggingFace sentence embeddings (all-MiniLM-L6-v2), a Chroma vector store, and Google’s Gemini language model over a curated seven-document clinical knowledge base spanning pain, digestive, neurological, respiratory, mental health, and medical history domains. It is deployed as a Streamlit web application, requiring no clinical infrastructure or proprietary patient data. Three experiments are conducted using a synthetic query set: retrieval quality assessment via Precision@k and recall across chunk sizes (256, 500, 1000 characters) and depths (k ∈ {2,3,5}); response quality evaluation using ROUGE-L and BERTScore against a non-retrieval baseline; and prompt engineering analysis comparing clinical, empathetic, and structured styles via an LLM-as-judge protocol. Results show that RAG consistently outperforms the baseline, with k = 3- and 500-character chunks yielding optimal retrieval performance, and empathetic prompts achieving the highest patient scores. These findings suggest that conversational AI can meaningfully improve pre-visit communication, with broader implications for reducing health disparities and supporting patients with limited health literacy in navigating complex healthcare systems.
Advisor
Mansi H. Bhavsar
Committee Member
Rushit Dave
Committee Member
Rajeev Bukralia
Date of Degree
2026
Language
english
Document Type
APP
Degree
Master of Science (MS)
Program of Study
Data Science
Department
Computer Information Science
College
Science, Engineering and Technology
Recommended Citation
Umuhoza, R. (2026). Previsit AI: A retrieval-augmented generation for patient readiness in clinical encounters [Master’s alternative plan paper, Minnesota State University, Mankato]. Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University, Mankato. https://cornerstone.lib.mnsu.edu/etds/1625/
Included in
Artificial Intelligence and Robotics Commons, Biomedical Informatics Commons, Data Science Commons
Rights Statement
