In partial fulfillment of the requirements for the degree of 

Doctor of Philosophy in Bioinformatics

in the School of Biological Sciences

Yuanbo (Cody) Wang

Defends his thesis:

Building a Systematic Analytic Pipeline – Big Data Innovation in Healthcare

Monday, August 5th, 2019

1:00 PM Eastern Time

BME/Whitaker Building, Room 1103

Thesis Advisor:

Dr. Eva Lee

School of Industrial and Systems Engineering

Georgia Institute of Technology

Committee Members:

Dr. King Jordan

School of Biological Sciences

Georgia Institute of Technology

Dr. Fredrik Vannberg
School of Biological Sciences

Georgia Institute of Technology

Dr. Yajun Mei

School of Industrial and Systems Engineering

Georgia Institute of Technology

Dr. Alfred Merrill

School of Biological Sciences

Georgia Institute of Technology

Dr. Shatavia Morrison

Division of Bacterial Diseases

Centers for Disease Control and Prevention

Abstract

Data-driven healthcare utilizing big data in Electronic Health Records (EHR) has the potential to revolutionize care delivery while reducing costs. However, for researchers, policymakers, and practitioners to take full advantage of the benefits that EHR can provide, several challenges must be addressed: 1) Extraction and coding methods for EHR data must be strategically designed to address issues of data quantity, quality, and patient confidentiality; 2) Standardization of clinical terminologies is essential in facilitating interoperability among EHR systems and allows for multi-site comparative effectiveness studies; 3) Effective methods for mining longitudinal health data common in the EHR are critical for revealing disease progression, treatment patterns, and patient similarities, each of which plays an important role in clinical decision support and treatment improvement; 4) Advanced machine learning techniques are necessary for early detection and prognosis of disease and identifying critical factors that impact patient outcome and; 5) Practical intervention strategies must be developed to address healthcare disparity in rural and remote areas with lack of resources and access. My thesis focuses on these five issues by developing a systematic analytic pipeline for big data in healthcare. Specifically, innovative strategies are developed for information extraction, clinical terminology mapping, time-series mining and clustering, feature selection and discriminatory modeling. Finally, practical implementation methods for telehealth services are designed to reduce healthcare disparity in underserved rural Appalachian counties in Georgia.