Health Informatics Research

Selected Research Areas of George Mason Health Informatics Faculty

Health informatics faculty at George Mason University conduct original research in several areas related to health informatics, health information technology, and health services research. Particular research areas of interest are electronic/personal medical records, intelligent systems, health care terminologies, data and text mining, consumer health informatics, clinical decision support, and health data privacy and security. Below are brief descriptions of some of these research areas. For details or information about our expertise in other areas, please contact the department.

Electronic health records, personal health records, and electronic medical records are interrelated areas of research investigated by our faculty. The research spans usability, interoperability, and integration of systems. Specific areas include decision support and intelligent capabilities of systems.

Health Data Analytics

Contact
Dr. Farrokh Alemi
Dr. Phan Giang
Dr. Janusz Wojtusiak

Data analytics is the future of health and health care. It is no longer possible for people to navigate the complexities of health and medical disciplines without data analytics. Our faculty interests span a wide range of topics including comparative effectiveness, causality, knowledge discovery from data, predictive modeling, and anomaly detection. In data analytic efforts, we closely collaborate with experts in health services research, health administration, and health policy.

Data Mining and Machine Learning in Health Care

Contact
Dr. Janusz Wojtusiak
Dr. Farrokh Alemi

Data mining is a process of discovering new interesting patterns and regularities in data. Data mining methods applied in health care help clinicians and administrators extend their knowledge and help decision making by sifting through large amounts of data. Machine learning is a related field that concerns creating learning capabilities in computers. By observing the environment, interacting with users, or analyzing past data, computer algorithms learn how to solve problems and aid decision makers by providing decision support.

Ontology-based Data Integration

Contact
Dr. Hua Min

The ability to resolve semantic conflicts between heterogeneous information systems is one of the major challenges in the data integration field. Our research aims to achieve semantic interoperability across systems using an ontology-based system.

Consumer Health Informatics and Patient Generated Data

Dr. Sorina Madison
Dr. Hua Min
Dr. Farrokh Alemi
Dr. Janusz Wojtusiak

The focus of this project is to understand consumer health communities, as well patient generated data and its uses within and outside healthcare systems. This topic spans a wide range of issues from policy, privacy and security, through potential uses and requirements, to technical issues in Big Data analytics.

Technology Adoption and Health Care Management

Dr. Sorina Madison

Research in this area seeks to understand the factors that affect technology adoption. Here the focus is mainly on health care organizations and the users, mostly health care professionals, of new health information technologies. The research draws from different fields such as communication science and management to understand factors that impede technology adoption in the health care industry.

Text Mining and Natural Language Processing in Health Care

Contact
Dr. Phan Giang

Clinical notes hold valuable information about patients and healthcare. Health professionals prefer clinical notes because of tradition and the advantages of a natural language including convenience, flexibility, and richness. Those properties of natural language create challenges for machines that process clinical notes. The research in clinical text mining focuses on: 1) the problems of semantic similarity for information extraction, and 2) translating narrative texts into structured format using standard vocabularies including HL7 CDA.

Decision Support

Contact
Dr. Farrokh Alemi
Dr. Phan Giang

Research focuses on decision models under mixed uncertainty: risk, ambiguity and ignorance. For many complex decisions especially in health care, the quantification of relevant uncertainty in terms of probability is unreliable and even impossible. In such situations, recommendations from traditional decision models using probability are not accepted by decision makers. Our research aims to develop a unified model for decisions under uncertainty that can deal not only with risk but also with ambiguity and ignorance that decision makers encounter in their practice.