Community Learning Space

The infraNET Project and 
Waterloo Institute for Health Informatics Research
at the University of Waterloo

present ...

Mining Health-Related Data:
Methods and Applications in Research, 
Public Health, and Patient Care


John Holmes, PhD
Assistant Professor, Medical Informatics in Epidemiology
Hospital of the University of Pennsylvania

Wednesday, October  22, 2003
3:00 p.m. - 4:30 p.m.

Davis Centre, Room 1302
University of Waterloo

This Seminar is Sponsored By 
    McKesson Canada 

This seminar is of interest to Health and IT Executives, IS/IT Staff, Faculty and Students.
There is no charge for this event, however, we ask that you register to attend.
Seating is on a first-come, first-served basis.

Presentation Archive 2003-2004


Health-related data is extremely rich in discoverable knowledge. In research domains, data are collected on everything from proteins and genes to animals and humans. In public health, surveillance systems capture data regarding the population at large. In patient care, ever-increasing amounts of clinical data are becoming the norm in both inpatient and outpatient settings. How do we make sense of these data and is there a way we can bring these diverse data systems together into an analyzable whole?

This talk examines approaches to warehousing these data and analyzing them using knowledge discovery tools. Applying these data mining tools is a process with a well-defined life-cycle. The talk will focus on this knowledge discovery life-cycle, with a review of each of the component articulations, from data aggregation and reduction, to data representation and visualization, to selecting a suite of data mining tools, and through to the interpretation of results. Three test cases, one each from research, public health, and patient care domains, will be used to illustrate these articulations. In addition, several data mining tools will be discussed and demonstrated.

John Holmes is Assistant Professor of Medical Informatics in Epidemiology at the Hospital of the University of Pennsylvania and a researcher at the Center for Clinical Epidemiology and Biostatistics at the University of Pennsylvania. He is also the Associate Director of the Post-Doctoral Fellowship Program in Medical Informatics at the Veteran s Administration Medical Center in Philadelphia. Dr. Holmes is a member in the Ackoff Center for Advancement of Systems Approaches, a community of researchers at the University of Pennsylvania working on a wide variety of projects in complex systems, medical informatics, and decision making.

Dr. Holmes obtained his PhD (Information Science) from Drexel in 1996. His dissertation, "Evolution-Assisted Discovery of Sentinel Features in Epidemiologic Surveillance, was the first work to apply evolutionary computation to epidemiologic research. Today, Dr. Holmes has an international reputation in applying evolutionary computation to epidemiologic data mining and has broadened his research interests to include several other areas in medical informatics including machine learning approaches to knowledge discovery in databases.

For more information

Shirley Fenton
Managing Director, WIHIR
The infraNET Project
Computer Systems Group, University of Waterloo
(519) 888-4074

Seminar Hosts

This seminar is hosted by the Waterloo Institute for Health Informatics Research (WIHIR) and The infraNET Project, University of Waterloo.

The infraNET Project, initiated by the University of Waterloo in 1996, is a partnership to advance Web and Internet technologies. Its founding partners are: LivePage (now part of Siebel), MKS, Open Text, RIM, Sybase (Waterloo) and Waterloo Maple.

We also gratefully acknowledge the assistance of the Institute for Computer Research, University of Waterloo.