The research interests of Richard (Rick) D. White, M.D., focus on the initial technical development, implementation, validation and eventual broad-scale clinical application of advanced magnetic resonance imaging (MRI) and computerized tomography (CT) techniques for assessment of cardiovascular diseases. This imaging is especially important in diseases such as heart failure and diastolic dysfunction, where improved understanding of the specific pathophysiological or genetic bases are needed.
Recently, Dr. White and colleagues have been pursuing a strong interest in deep learning (DL), machine learning (ML) and artificial intelligence (AI). These technologies have great potential to enhance efficiency and effectiveness in delivering patient care, as well as to improve understanding of disease processes and their avoidance or treatment.
- Initial technical development, implementation, validation and eventual broad-scale clinical application of advanced MRI and CT techniques for assessment of cardiovascular diseases
- Promotion of patient-specific integrated multimodality imaging such as radiography, MRI and CT for the assessment of cardiovascular diseases
- Promotion of radiomics to precision imaging toward the concept of patient-specific indirect image-based biopsies as an alternative to invasive direct tissue-biopsy procedures
- Promotion of clinical applications of DL, ML and AI in imaging, including improved safety, efficiency and effectiveness
- Identification and optimization of standard operating procedures in the life cycle of AI model development to ensure representation, meaningful beneficial use and avoidance of harm
Significance to patient care
Non-evidence-based, inappropriate use of "noninvasive" imaging (including radiography, MRI and CT) leads to unwarranted exposure of patients to risks from ionizing radiation, high or rapidly changing electromagnetic fields, or contrast media. Even when imaging is justified based on evidence and clinical need, the specific details of each radiologic examination must be recorded as a routine informatics exercise.
Next maximal image data use in clinical care is predicated on its prior reconstruction and storage, retrieval and display, post-processing and evaluation, and reporting and integration with other patient data types (such as laboratory results). All the aforementioned parameters apply to multimodality noninvasive cardiovascular imaging, with DL, ML and AI potentially assisting the physician in achieving the most efficient and effective delivery of patient care.
- Chairman and chief of service, Department of Radiology, Ohio State University College of Medicine, 2010-2020
- Recipient, Distinguished Thesis Award for Health Informatics, "Artificial Intelligence-Augmented Exclusion of Coronary Atherosclerosis on CCTA for Chest Pain: Initial Performance and Technical Issues," Northwestern University School of Professional Studies, 2019
- Ad hoc program reviewer, Biomedical and Metabolic Imaging Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, 2013
- Chairman, Appropriateness Criteria Committee: Expert Panel on Cardiac Imaging, American College of Radiology, 2008-2012
- ACRIN midterm reviewer, Cancer Imaging Program-NCI: National Institutes of Health, 2011
- Chairman, National Center-Council on Cardiovascular Radiology & Intervention Committee on Cardiovascular Imaging and Intervention, American Heart Association, 2007-2010