Each day, in hospitals across the country, patients experience life-threatening events. Their survival often depends on the fast reaction of the experienced care providers who watch constantly over them. But all too often, care staff have very little warning of impending deterioration. By the time overt symptoms appear, the damage has already been done. My research focuses on developing computer algorithms which can be used to constantly watch over patients, and provide early warnings of life-threatening events. This is possible because acute decomposition does not manifest instantaneously. It builds up over a short period of time, changing the physiologic dynamics of the patient in the process. These precursors can be subtle and hard for a human to see, given the level of noise in patient data, but a computer can be trained to detect such patterns. By using technology to anticipate such events in real-time, care can be provided proactively instead of reactively, which has tremendous potential for improving survival in critically ill patients. This represents a fundamental shift in the way that traditional patient monitoring and surveillance has been conducted over the last 50 years.
Publications/Creative Works
Click here to search for this faculty member's publications on PubMed.
Affiliations
Training Grants
NLM Training Program in Biomedical Informatics & Data Science for Predoctoral and Postdoctoral Fellows
Important Disclaimer: The responsibility for the accuracy of the information contained on these pages lies with the authors and user providing such information.