Christopher R. Genovese
Can I use statistics to learn what you are thinking? Well... not quite, but I can use statistics to learn how you are thinking. For the past two years, I have been working in an exciting new research area, called functional neuroimaging. This involves using state-of-the-art technology to acquire images of the human brain while it is actively engaged in cognitive processing. Functional neuroimaging is a truly multi-disciplinary area; I work with physicists, psychologists, neurologists, radiologists, engineers, and computer scientists to solve a wide range of interesting problems.
I have done most of my work using a particular neuroimaging technique, functional Magnetic Resonance Imaging or fMRI. During an fMRI experiment, three-dimensional Magnetic Resonance images of the subject's brain are acquired while the subject performs a sequence of tasks (e.g., reading sentences, remembering and recalling lists, etc) that are designed to exercise specific cognitive processes. The sequence of images so acquired forms a four-dimensional data set that contains critical information about how the cognitive processing is performed in the brain. The concentrated firing of a cluster of neurons in the brain leads to a small physiological change that is (barely) detectable in the measured fMRI data. Scientists can thus use fMRI data to help identify the neural processes underlying cognition and to build and test theoretical models of how the brain works. But drawing scientifically useful inferences from such complex data is fraught with difficult statistical challenges.
Statistics is a unique field in that what we do is relevant to almost every scientific discipline. What I find so enjoyable about statistics is that it gives us the tools to have a scientific impact in so many directions. In functional neuroimaging, I collaborate with neuroscientists to tease out the workings of the brain and with clinical specialists using fMRI to improve diagnosis of stroke and understanding of schizophrenia. I also collaborate with astrophysicists who are trying to infer the internal structure of the Sun---and thus the basic structure of all stars---from observations of its surface oscillations. I collaborate with solid state physicists who seek to understand the dynamic properties of particle systems and complex biological molecules. I collaborate with mathematicians to clarify the abstract properties of certain high-dimensional statistical problems. And that's only an average day...
Some Related Publications
Genovese, C.R., Stark, P.B., and Thompson, M.J. (1995). "Uncertainties for Two-Dimensional Models of Solar Rotation from Helioseismic Eigenfrequency Splitting," Astrophysical Journal, 443, 843-854.
Genovese, C.R. (2000). "A Bayesian Time-Course Model for Functional Magnetic Resonance Imaging Data," Journal of the American Statistical Association, to appear.
Genovese, C.R. and Sweeney, J.A. (1999). "Functional Connectivity in the Cortical Regions Subserving Eye Movements (with discussion)," in Case Studies in Bayesian Statistics, 4, eds. Kass, R.E., Carlin, B.P., Carriquiry, A.L., Gatsonis, C., Gelman, A., Verdinelli, I., and West, M., Springer-Verlag, 59-132.