SEMINAR:

"Personalized Medicine: Are We There Yet?"

Prof. Lucila Ohno-Machado, Harvard Medical School

When: 09 Ottobre 2007 - h. 11.00

Where: Parco tecnologico di Pula - Edificio 2

Il seminario è aperto a tutti.

Seminar is open to all - everybody welcome.

Abstract
The advent of new measurement technologies for biological phenomena and whole genome scans has fueled much discussion on the potential of personalized medicine. To personalize medical care, preventive measures and therapies will need to be tailored to individuals, instead of what happens today, when interventions are primarily tailored to diseases. As technology gets cheaper and more accurate, obtaining extensive genotypic information for large numbers of individuals becomes economically feasible. However, current data and data analyses methods are still inadequate to handle this information. Even with all the new data, the relationship between genotype and disease is likely to remain non-deterministic, as unknown environmental factors are likely to continue to play a significant role in health and disease.
Can pattern recognition at the individual and disease levels be combined in models that recommend individualized prevention strategies or personalized therapies?
In this talk, I will discuss this question from the perspective of evaluating the quality of individualized estimated produced by predictive models, since these estimates are often used to “individualize” interventions. Special emphasis will be given on calibration aspects of different types of predictive models used in medical practice. Examples from different medical specialties will be discussed to illustrate the issue.

Speaker
Lucila Ohno-Machado has received her PhD in medical information sciences and computer science from Stanford University, and her medical doctor degree from the University of Sao Paulo. She is associate professor of radiology and directs the Decision Systems Group at Brigham and Women’s Hospital, Harvard Medical School, as well as the Boston Training Program in Biomedical Informatics. She teaches courses in artificial intelligence in medicine and biomedical decision support at Harvard-MIT and is interested in the development of new algorithms for development and evaluation of predictive models, with particular emphasis on calibration.

Research - Master Course and Seminar Page