NLM’s health information standards and discovery research focuses on the development of methods to gain insights from large health databases while learning the strengths and weaknesses of datasets and improving them, when possible. This area of research assesses whether specific standards are fit for purpose (e.g., quality assurance and interoperability assessments of biomedical terminologies) and investigates standards in action (e.g., in support of tasks such as natural language processing, annotation, data integration, and mapping across terminologies).
Acting Director, LHNCBC, Senior Investigator, Computational Health Research Branch
Olivier Bodenreider, MD, PhD
- Conducting health information standards research with a focus on ontologies and terminologies
- Assessing whether specific standards are fit for purpose (e.g., quality assurance of SNOMED CT, interoperability between RxNorm and SNOMED CT drugs, assessment of ICD-11)
- Investigating new methods for aligning ontologies (e.g., deep learning approaches to assessing synonymy in the Unified Medical Language System Metathesaurus)
Senior Investigator, Office of the Director, NLM
Clem McDonald, MD
- Leveraging observational data (ERH, claims) for research purposes
- Comparing the effects of commonly used classes of blood pressure, diabetes, and lipid-lowering medications (in 300,000 Medicare-aged diabetic patients)
- Investigating the risk of tendon rupture for the five most commonly prescribed oral antibiotics (for Medicare)
- Investigating the risk of proton pump inhibitors for survival (in 2.3 million Medicare patients)
Independent Investigator, Computational Health Research Branch, LHNCBC
Jeremy Weiss, PhD, MD
- Leads the Care Health and Reasoning Machines (CHARM) lab to advance machine learning methods in clinical data science.
- Focuses on improving clinical prediction in internal medicine and critical care settings, with applications in opioid abuse, sepsis, and COVID-19.
