Theses and Dissertations

ORCID

https://orcid.org/0009-0006-5492-8113

Advisor

Rahimi, Shahram

Committee Member

Perkins, Andy

Committee Member

Mittal, Sudip

Committee Member

Chen, Zhiqian

Date of Degree

5-16-2025

Original embargo terms

Visible MSU Only 6 months

Document Type

Dissertation - Campus Access Only

Major

Computational Engineering

Degree Name

Doctor of Philosophy (Ph.D.)

College

James Worth Bagley College of Engineering

Department

Computational Engineering Program

Abstract

Healthcare advancements require advanced solutions to process and understand complex patient data. This dissertation outlines a detailed framework for mapping patient journeys with knowledge graphs (KGs), which begins by analyzing existing Patient-Centric Knowledge Graphs (PCKGs) literature to identify current research gaps in the representation and analysis of patient journeys. Our research proposes solutions to these gaps through three main contributions. First, we introduce the Patient Journey Ontology (PJO), which enables systematic encoding of patient encounters, diagnoses, treatments, and outcomes into a semantic knowledge structure designed for interoperability. The construction of patient journey knowledge graphs (PJKGs) is based on this foundational ontology. Building upon this ontology, we present an innovative framework to construct PJKGs automatically through Large Language Models (LLMs) that convert clinical dialogues into structured knowledge representations that potentially allow tracking of patients’ entire medical journeys. Furthermore, we propose the Dynamic Feature and Temporal Similarity (DFTS) framework, a hybrid approach that combines feature-based and temporal-based KG similarity methods with dynamic weighting mechanisms. Unlike traditional machine learning approaches that require large datasets, DFTS is designed to work effectively with limited healthcare data while maintaining scalability for growing datasets. A case study on chronic disease management demonstrates the effectiveness of the proposed framework through its ability to identify patients with similar medical journeys. The findings of this dissertation demonstrate the potential of structured knowledge representation and analysis to enhance healthcare decision-making support systems while also advancing patient-focused care strategies. By providing a comprehensive framework for mapping, constructing, and analyzing patient journeys, this work establishes a foundation for more effective and personalized healthcare delivery.

Download

Share

COinS