Sunghwan Sohn, Ph.D., has expertise in mining large-scale electronic health records (EHRs) to unlock unstructured and hidden information using natural language processing and machine learning. Dr. Sohn's research goal is to create new capacities for clinical research and practice in order to achieve better patient solutions.
- Information extraction and normalization. Dr. Sohn applies natural language processing techniques to extract unstructured medical concepts from clinical narratives and map them to standard forms in order to facilitate clinical research and information exchange across health care institutions.
- Patient cohort identification. Dr. Sohn develops algorithms to identify patient cohorts with specific medical conditions such as peripheral artery disease and abdominal artery aneurysm by mining through various EHRs to enable a large-scale epidemiological study.
- Automated chart review. Dr. Sohn conducts research in developing algorithms and systems that can automate manual chart review of asthma ascertainment based on multiple criteria to facilitate population-based asthma research.
- Surveillance of post-surgical complications. Dr. Sohn is interested in mining and analyzing various types of EHRs to detect and predict surgical site infections and bleeding after colorectal surgery in near-real time to support clinician decisions for better treatment and outcome.
Significance to patient care
Dr. Sohn's research facilitates the best use of EHRs to solve clinical problems and improve public health. His work provides biomedical scientists and clinicians access to unstructured information from clinical narratives and clinical text analytics necessary for clinical research and patient care.