U of L professor Somnath Datta developing model to explore the secrets of randomness
A University of Louisville professor is on a quest to explore the secrets of randomness, developing a model that can be used for everything from predicting a person’s health prognosis to, apparently, national security.
Dr. Somnath Datta, a mathematician and professor in U of L’s Department of Bioinformatics and Biostatistics, School of Public Health and Information Services, has received grants to develop statistical models describing how one or more random variables relate to other variables.
The total value of the grants, which have a renewal clause, could about $74,000.
Statistical models can be useful or inconclusive.
Datta is on a quest to build a multi-state statistical model that could provide more accurate medical information. In other words, be useful. A multi-state statistical model examines process – for instance, how a person moves from one stage of a disease to another over time.
For cancer patients, this model could analyze how the disease progresses from one stage to another, or determine what prolongs cancer free survival for those in remission.
Datta recently received a two year grant from the National Security Agency through their Mathematical Sciences Program to employ a multi-state statistical model that will incorporate both medical and genomic data.
“The model will be derived from empirical evidence – based on observation or experience – rather than unverifiable mathematical laws,” Datta said. “As a result, the prediction from the methodology will be more robust and less likely to include model misspecification errors.”
In addition to having applications to cancer research, Datta’s new statistical method might be applicable to many other disciplines – ranging from engineering to political science – that deal with staged systems, any systems that can move from one phase or stage to another.
In politics, for instance, this would involve all the elements from planning a campaign to taking office.
Datta has explored a wide variety of subjects during his career.
In addition to medical research, he has created statistical models that predict anthrax terrorism probabilities, airline economics, and marketing patterns.
Datta and his team won’t have to compile the data bases that will fuel their work.
They’re everywhere. From governments to industries, data bases are widely available.
For instance, every time you see a U.S. weather forecast, it was built from data produced by the National Oceanographic and Atmospheric Administration.
Datta will begin preparing a grant request this year to the National Institutes of Health to create a statistical model that will employ an even larger number of data bases.