Colloquium - Richard Martin

In this talk I describe research to improve the workflow of an

Emergency Department (ED) using Real Time Location Systems.  Our work

advances the state of the art of clinical workflow by automatically

inferring where patients are in a ED task flow graph from

the spatial-temporal location, proximity and mobility events of care

providers, patients and equipment. Our Bayesian approach provides the

likelihoods of clinical tasks with high true positive and low false

negative rates. A second part of this work quantifies the impact of

two different classes of actions care providers can perform using

automated workflow information: dynamic staffing re-assignment and

increased accountability.

Bio:

Dr. Richard Martin is an associate professor at Rutgers University.

His current research interests include wireless device localization,

and human factors in dependable computing.  He is a member of the

Wireless Network Information Laboratory (WINLAB) an co-leads the PANIC

(Pervasive Available No-futz Internet Computing) Laboratory.  His

awards include the best paper award at the 2004 IEEE Conference on

Sensor and Ad Hoc Communication Networks as well as a CAREER award

from the National Science Foundation.  Dr. Martin has served as an

investigator on grants from the Defense Advanced Research Projects

Agency, the National Science Foundation, and IBM. He received a

B.A. from Rutgers University and an, M.S. and Ph.D. in computer