MACHINE LEARNING THAT MAKES SENSE IN CLINICAL SETTINGS - A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

Aug 3, 2026

This dissertation presents a health services research study focused on how machine learning methods can be meaningfully applied within clinical environments. It introduces the context, motivations, problem statement, and research questions, then develops the dissertation’s structure through an experimental and theoretical foundation. The work reviews major machine learning paradigms, including supervised, unsupervised, semi-supervised learning, and neural approaches, and connects them to medical use cases such as electronic health records, clinical decision support, uncertainty, probabilistic reasoning, and ICU risk prediction and severity scoring. It also addresses adoption challenges for implementation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages200

Reading Time8 h 24 min

AccessFree to download

Introduction

Context

Motivations

Problem statement

Research Questions and Dissertation Aims

Experiment 1

Contributions and Deliverables

Overview of Dissertation and Roadmap

Theoretical Background

Overview of Machine Learning

Types of Machine Learning

Supervised Learning

Unsupervised Learning

Dimensions Reduction

Clustering

Association

Semi-Supervised Learning

Artificial Neural Networks

Reinforcement Learning

Temporal Difference Learning

Machine Learning in Medicine

Electronic Health Records

Clinical Decision Support Systems (CDSS)

Medical Decision Making

Uncertainty

Probabilistic Medical Reasoning

Machine Learning in ICU

Risk Prediction and Severity Scores in ICU

Adoption and Challenges of ML Implementation in ICU

Review of Characteristics of ML Models in Clinical Field

Guidelines/Framework on Building Models in Clinical Settings

MACHINE LEARNING THAT MAKES SENSE IN CLINICAL SETTINGS by Eman Elashkar A Dissertation Submitted to the Graduate Faculty of George Mason University in Partial Fulfillment of The Requirements for the Degree of Doctor of Philosophy Health Services Research Committee: Dr. Janusz Wojtusiak, PhD Chair GMU, Health Administration and Policy Dr. Hua Min, PhD __________________________________ Committee Member GMU, Health Administration and Policy Dr. Sanja Avramovic, PhD Committee Member GMU, Health Administration and Policy . Dr. Gilbert Gimm, PhD Program Director Dr. PJ Maddox, Department Chair Date: Summer Semester 2024 George Mason University Fairfax, VA MACHINE LEARNING THAT MAKES SENSE IN CLINICAL SETTINGS A Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy By Eman Elashkar Bachelor of Medicine and Bachelor of Surgery Kasr Al Aini School of Medicine, Cairo University Director: Janusz Wojtusiak, Professor College of Public Health, George Mason University Summer Semester 2024 George Mason University Fairfax, VA Copyright 2024 Eman Elashkar All Rights Reserved DEDICATION To my family, whose unwavering love and support have been my bedrock; to my parents, whose encouragement, and belief in me have been my greatest strength; to my friends, whose companionship have provided me with joy and motivation throughout this process, and to my esteemed colleagues, whose collaboration and insights have been invaluable in pursuing this work. With heartfelt gratitude and deep appreciation, this dissertation is dedicated to all of you. ACKNOWLEDGMENTS I wish to thank my advisor, Dr. Janusz Wojtusiak, for his guidance and support throughout this process and my whole health informatics career. I also want to thank the members of my committee, Dr. Hua Min and Dr. Sanja Avramovic for their help and support. Finally, my work would not have been possible without the support of my colleagues. TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. xi LIST OF FIGURES .......................................................................................................... xii LIST OF EQUATIONS ................................................................................................... xiv LIST OF ABBREVIATIONS AND SYMBOLS ............................................................. xv ABSTRACT..................................................................................................................... xvi INTRODUCTION .............................................................................................................. 1 Context ................................................................................................................ 1 Motivations.......................................................................................................... 3 Problem statement ............................................................................................... 5 Research Questions and Dissertation Aims ........................................................ 8 EXPERIMENT 1 ........................................................................................... 13 Contributions and Deliverables ......................................................................... 15 Overview of Dissertation and Roadmap ........................................................... 16 THEORETICAL BACKGROUND.................................................................................. 18 Overview of Machine Learning ........................................................................ 18 Types of Machine Learning .............................................................................. 19 SUPERVISED LEARNING..............................................

FAQs

What is the dissertation’s main focus in clinical settings?

It examines how machine learning can be applied in meaningful ways within clinical environments, with emphasis on medical decision support and patient-risk related tasks.

Which machine learning approaches are covered in the theoretical background?

The document reviews supervised, unsupervised, semi-supervised learning, as well as topics such as dimensionality reduction, clustering, association, artificial neural networks, reinforcement learning, and temporal difference learning.

How does the dissertation connect machine learning to healthcare workflows?

It discusses applications such as electronic health records, clinical decision support systems, uncertainty handling, probabilistic medical reasoning, and specific ICU risk prediction and severity scoring tasks, along with adoption challenges.

Aug 3, 2026

Upload by River Wang