New User? Sign Up Free
SCCM is performing maintenance on its websites. For the best browsing experience, please use Microsoft Edge or Safari. Those using Chrome or Firefox may experience access issues at this time.
Matthew Churpek, MD, MPH, PhD, will share the top 10 things he wished someone had told him about clinical predictive modeling as part of the thought leader session Data Science and Critical Care at the 2022 Critical Care Congress.
As an associate professor of medicine in the Division of Pulmonary and Critical Care and affiliate faculty in the Department of Biostatistics and Medical Informatics at the University of Wisconsin Madison, Matthew Churpek, MD, MPH, PhD, spends about 20% of his time as an attending physician in the intensive care unit (ICU). Most of his time is dedicated to leading the school’s ICU Data Science Lab, where he focuses on developing and implementing machine learning algorithms for critical illness. The lab uses electronic health record data along with epidemiology, biostatistics, and machine learning methods to improve the care of hospitalized patients.
Dr. Churpek will share the top 10 things he wished someone had told him about clinical predictive modeling as part of the thought leader session Data Science and Critical Care at the 2022 Critical Care Congress. The virtual discussion will be held Thursday, April 21, 2022. Dr. Churpek will present along with Gilles Clermont, MD, MS.
“Data science and machine learning are already a part of our everyday lives. Over the next several years it will become more common in clinical care as well,” Dr. Churpek said. “Educating and increasing the interest of clinicians and researchers about the field will help them better adapt to these technologies in the future, be informed decision-makers related to how and when to use these algorithms at the bedside, get involved in this area at the local and national level, and create new collaborations and partnerships across disciplines, which ultimately will help our patients.”
Dr. Churpek first realized the potential machine learning can have in healthcare when he saw patients in the ICU with a critical illness or cardiac arrest. He thought an algorithm should be able to identify high-risk patients before cardiac arrest occurred and hoped it would lead to better patient outcomes. He reasoned that, if algorithms can sort email, recommend music, and protect people from fraud, maybe an algorithm could help critically ill patients as well.
It is this type of questioning that Dr. Churpek believes will be key for healthcare professionals to consider if they want to get started in the world of data science by identifying ways that patient care or clinician experiences can be improved with data or smarter algorithms.
“By looking closer at our own mistakes, experiences, and how we spend our time, we can discover ways to improve things,” he said. “After that, the next step is seeing if there is a researcher or analytics group who might be interested in collaborating on this problem. Translating data to models to implementation is a complex and challenging process, but with the right support, collaborators, and a multidisciplinary approach, you have the potential to improve care and patient outcomes.”
Posted: 3/15/2022 | 0 comments
Log in to Comment