PliaTech was founded originally as Pliable Software by Michael B. Pliam, MD, PhD
in 1999. Dr. Pliam has a long-standing interest in the statistical modeling
of cardiac surgical outcomes which he has been involved in for the past decade.
Given the present state of health care delivery in our country, it has become apparent
that we need to find improved statistical modeling techniques in order to
provide for a more solid foundation for the "evidence-based" practice of medicine.
As a primary prerequisite to that goal, it was necessary
to gain a deeper understanding of the underlying nature of current statistical models.
That requires more than an passing knowledge of statistics, modeling, probability
theory, and linear matrix algebra, Dr Pliam began writing customized software to
explore these fields. The result was a set of very specialized proprietary
tools that greatly enhances our group's ability to perform statistical modeling.
After a time, it was realized that these software
programs and applications might be of considerable use to others, most particularly
students of statistics, engineering, and mathematics. Consequently, we have
made some of our proprietary sofware available to others.
After several years of independent research and
study, PliaTech was created as a small, privately controlled
organization. The mission
was to find new and creative uses for data analysis to in the field of health care
to provide a more solid ground than currently existed for the practice of 'evidence-based'
medicine and surgery.
Most recently we have again become interested in Artificial Neural Networks
(ANN) which was very popular in the 1970's when enthusiasm for AI was
booming at MIT. In the 1990 s Canadian graduate student Colleen Ennett
wrote a Masters Thesis showing that an ANN could be used to predict outcomes
in cardiac surgery equal to or superior to that obtained using more
conventional methods of Bayesian or logistic regression modeling. We
are currently working to develop an ANN to predict hospital costs related to
disease severity in cardiac patients.