John F. Hurdle, MD, PhD

Main Research Interests: practical, real world natural language processing (NLP) for clinical and biomedical text applications. Past major research contributions: patient safety with a focus on adverse drug events; data mining in renal transplantation databases; more NLP, of course, and finding smart and inexpensive ways to assess, and relate to healthcare outcomes, the grocery quality of the foods households buy at grocery stores and online.

Education: Dr. John F. Hurdle earned his MD from the University of Colorado (1981) and his MS in Computer Science from Columbia University the same year, followed by a PhD in Computer Science from the University of Utah (1994). He took his informatics postdoctoral training here at Utah (1996-97). Following that training, Dr. Hurdle worked as a research clinical informaticist at the Salt Lake Veterans Medical Center’s (VAMC) Geriatric Research, Education, and Clinical Center (1998-2004). While there, he was the Principal Investigator of two VA Health Services Research and Development (HSR&D) Grants, the first HSR&D grants funded at the SLC VAMC in over two decades. His small center grant from HSR&D has grown into a multi-million dollar HSR&D enterprise at the VAMC. That work directly spurred the creation of the VA’s primary informatics research infrastructure, VINCI.

At the University of Utah: He joined the Department of Biomedical Informatics (BMI) as a regular faculty member in 2005. In the transition from the VA to the University he served as IRB Chair during the migration from paper records to an all-electronic IRB system called ERICA. In 2007 he served as a Senior Fellow at the National Library of Medicine. Dr. Hurdle holds the rank of Professor. His work in the BMI spans clinical informatics, clinical research informatics, and public health/consumer health informatics. In addition to a long-standing commitment to improving patient safety, he directed a lab that focuses on clinical natural language processing (NLP) and another lab that focused on nutrition data mining (NDM). The POET NLP lab built tools to unlock the content of clinical narratives using NLP, so that this content can be used to assist the healthcare enterprise. The QualMART NDM lab was comprised of University of Utah researchers, graduate students, and post-docs who have expertise in computer science, biomedical informatics, nutrition, and consumer behavior. Our long-term goal focused on improving the quality of what people eat in order to improve their overall quality of life. Our approach weaves together consumer behavior theory, high-performance computing and simulation, cutting-edge nutrition modeling, and large-scale database management.

Etc.: In addition to research, he is active in both service and education,. He served as the Department’s Director of Graduate Studies from 2008 to 2016 and was the principal investigator of the BMI T15 NLM Training Grant for nearly a decade. Dr. Hurdle has served as a grant reviewer for the National Library of Medicine’s Standing Study Section (2008-2012) and continues to participate in special emphasis panel grant reviews for NLM and other NIH institutes. He served as Chair of the American Medical Informatics Association’s Ethics Committee when it created AMIA’s first code of professional conduct. He also served as member, Vice Chair, and Chair of the University’s Institutional Review Board (1999 – 2016) as well as Chair of the Resource for Genetic and Epidemiologic Research (2011-present), the data governance body overseeing the use of the Utah Population Database.

“I was once asked by a recruiter how I could rationalize staying in academic biomedical informatics rather than work in industry for a lot more money. ‘Oh that’s easy’, I told her, ‘In industry I would be working on what they want – in academics, I work on what I want." That's worth a lot of money.

Research Statement

I started my informatics career with a keen desire to improve patient safety, especially in the realm of medication errors (which, at the time, were estimated to be causing over 100,000 needless fatalities a year in the US). Over the past 15 years, I developed a very strong interest in applying natural language processing (NLP) tools to clinical narratives. My work on adverse drug events convinced me that there were important signals in clinical notes (aka "unstructured data") that can be used to improve patient care. After completing an NLP Senior Fellowship sponsored by the National Library of Medicine, I was awarded two research grants from the NIH to explore the utility of preprocessing clinical text to make clinical NLP more useful and efficient. My POET natural language processing lab works to develop practical and efficient tools that can extract information locked in clinical narrative text. Increasingly we need the information that can only be found in text to augment traditional structured information like laboratory test results, or medication lists. In combination, these two types of data are essential for modern data analytics and data science apps.

More recently I developed an interest in the nascent field of Nutrition Informatics. So my other lab focuses on nutrition data mining (NDM). Nutrition informatics holds great promise, especially as the United States faces an epidemic of diet-related diseases such as obesity, type II diabetes, and malnutrition. Nutrition informatics is very much in its infancy, and our Department is the first major informatics program to engage nutrition in a serious and principled way. Our primary goal for this work centers on building scalable tools that can measure the food quality of what households buy (in grocery stores or online) coupled to targeted, personalized recommendations designed to nudge households towards a healthier household food environment.