The SRU computational physics research group is run by Dr. Herat and Dr. Valera. Dr. Herat's research focuses on the application of physical models and simulation techniques to material science, biological systems, and medical applications. He is currently working on a cross-disciplinary medical imaging research project. This investigation focuses on fMRI neuroimaging analysis to investigate brain activation in different cohorts of individuals. Dr. Herat frequently seeks to involve student research assistants in his research project. He often has research grant money to pay student stipends.
Mathematical modeling and computations are essential components of modern research in physics. Physicists have very precise mathematical theories describing how various systems behave. Unfortunately, very often the equations that describe these theories are so complex that solving them by hand is not realistic. This is where the computational physicists come in. With the use of contemporary computing technologies, they perform these complex and vast calculations (or simulations) that cannot be done using the traditional techniques. Areas of applications include condensed-matter physics, astrophysics, elementary particle physics, and medical physics.
Dr. Herat's research specialty is computational physics. In particular his research agenda focuses on the application of physical models and simulation techniques to material science, biological systems, and medical applications. He frequently involves undergraduate research assistants in his research projects, and currently mentors two students. Dr. Herat often has research grant money available to pay student stipends. Two of his recent projects are briefly described below:
A collaborative computational biophysics research project. The primary objective of this research effort is to
numerically investigate the dynamics of protein aggregation. Proteins are essential parts of organisms and play a key role in a wide variety of biological processes. However, proteins also have a propensity to misfold and aggregate, which not only alter their original biological function but also can be harmful to the organism. We are looking at a particular type of misfolding of proteins into hedgehog shaped structure (or Spherulites) that is known to cause number of neurodegenerative diseases. It is our hope that shedding light into this malformation of proteins could one day lead to cures for some of the most debilitating deceases, including Alzheimer's, Parkinson's, and Creutzfeldt-Jakob's decease.
He is also currently working on a cross-disciplinary medical imaging research project. This investigation focuses on fMRI neuroimaging .Functional Magnetic Resonance Imaging (fMRI) is a neuroimaging technique that measures brain activity during a specific experiment over a time period by detecting associated changes in the oxygenation of blood flow. fMRI studies can be used to produce brain activation maps showing which parts of the brain are active while people are thinking, feeling or performing cognitive tasks.
The existing statistical methods used in fMRI analysis have several limitations. There are certain characteristics of fMRI studies that contribute to these limitations. First, fMRI sample sizes are small. Another compounding characteristic is that the data is highly correlated. As a result, it is challenging to observe statistically significant brain activation. The currently used analysis technique does not lend itself to adequately handle these limitations. This study will evaluate a more novel analysis technique known as Partial Least Squares Regression (PLSR) to be used as the statistical tool in fMRI analysis. This technique uses a rigorous mathematical procedure called Eigen Decomposition, which directly addresses the issues posed by the small sample sizes and the highly correlated data encountered in these studies. Establishing PLSR as a reliable and valid method of analysis for fMRI data will advance the field of neuroimaging.
Dr. Valera's research focuses on studying Soft Condensed Matter Systems using computational techniques. For the past six years I have been involved in work using Molecular Dynamics to simulate systems such as colloids and water.
Colloid research is a field of critical importance for the development of a wide range of products and industrial processes. Examples of colloidal suspensions include inks, paints, lubricants, cosmetics and milk, products that are ubiquitous in everyday life. One of the most important features is that several of the properties of these systems can be changed by exposing them to external factors, such as electric or magnetic fields. In this way we can control or "tune" the physical behavior of the system and this could lead to a new set colloidal structures. In addition, the tunability of the structures allows for the possibility of creating advance materials whose functions and properties can be switched on and off at will. These features of colloidal suspensions give rise to a range of interesting materials that includes ferrofluids. Using computer simulations, we are able to study ferrofluids under different constraints. We look for properties such as glassy states and/or organized behavior.