Assistant Professor, Computer Science Department
227 Hackerman Hall
Johns Hopkins University
3400 N. Charles St.
Baltimore, MD 21218
Phone: (410) 516 4126
E-mail: ssaria | at | cs.jhu.edu
Our healthcare budget is nearing three trillion dollars. Yet, the US ranks below most developed countries in terms of health outcomes. One of the largest under explored avenues is the better use of information derived from the vast amount of health data now being collected digitally.
At a very high-level, my interests lie in developing data frameworks that can leverage this routinely collected clinical data for improving the delivery of health care and reducing healthcare costs.
For example, in the neonatal ICU, my work shows that computational markers derived from routinely collected physiologic data at the bedside can predict infants who are at risk for major complications downstream. This led to a novel, automated, and low-cost approach for risk stratification and triage in preemies called the Physiscore (akin to the Apgar score).
Two current example thrusts in my lab include: developing computational methods for providing tailored treatment recommendations for patients with complex, chronic conditions; and developing methods for detecting adverse events in the critical care setting.
My works utilities ideas from machine learning and statistics to learn models from clinical data that are often riddled with noise and bias. Moreover, these data are often unstructured, high-dimensional, and noisy and the useful signals are often subtle and buried so we develop methods by which we encode plausible biological priors to aid discovery of informative representations from such data.
Bates D. W., S. Saria, L. Ohno-Machado, A. Shah and G. Escobar (2014). "Big data in health care: using analytics to identify and manage high-risk and high-cost patients." Health Aff (Millwood), 33(7): 1123-31.
Wayock C. P., R. L. Meserole, S. Saria, J. M. Jennings, T. A. Huisman, F. J. Northington and E. M. Graham (2014). "Perinatal risk factors for severe injury in neonates treated with whole-body hypothermia for encephalopathy." Am J Obstet Gynecol, 211(1): 41 e1-8.
Saria S. (2014). "A $3 Trillion Challenge to Computational Scientists: Transforming Healthcare Delivery." Intelligent Systems, IEEE, 29(4): 82-87.
Zimolzak A. J., C. M. Spettell, J. Fernandes, V. A. Fusaro, N. P. Palmer, S. Saria, I. S. Kohane, M. A. Jonikas and K. D. Mandl (2013). "Early detection of poor adherers to statins: applying individualized surveillance to pay for performance." PLoS One, 8(11): e79611.
Paxton C., A. Niculescu-Mizil and S. Saria (2013). "Developing predictive models using electronic medical records: challenges and pitfalls." AMIA Annu Symp Proc, 2013: 1109-15.