Skip to main content About News Giving All Departments Contact Us Site Map
 University of Texas Southwestern Medical School
 
Search       
Print Friendly  
spacer Home Education Research Patient Care Faculty & Administration Resource Careers
Faculty Directory Administration Administrative Departments
border=0
| Home > Faculty & Administration >
Patrick Carmack

 
 
Faculty Directory
 
 
Find a Doctor
 
 
Faculty Research Interests
 
 
Search Help
 
 
Update Faculty ProfileAccessible on campus or vpn
 
 
 

Patrick Carmack, Ph.D.

 Details of Research

Biographical Sketch Details of Research Personal Overview How to Contact
Patrick Carmack
Name:
  Patrick S Carmack, Ph.D.
Academic Title:
  Assistant Professor
Primary Appointment:
  Clinical Sciences
Secondary Appointment:
  Internal Medicine
School:
  Southwestern Medical School
Department Website:
  Department of Clinical Sciences
Email:
  Patrick Carmack, Ph.D.

 RESEARCH OVERVIEW
 
I am investigating various ways to improve statistical methodologies in neuro imaging (fMRI, ASL, PET, SPECT). This includes improvements in image pre-processing, brain mapping, and analysis techniques. The multiple comparisons problems plagues most current methods since there are literally hundreds of thousands of statistical tests. My general approach is to greatly reduce the number of comparisons by reducing the number of tests several orders of magnitude using spatial modeling (kriging) to obtain independent data summaries. I have also published a specific brain mapping improvement in a standardized space and am currently working on a project that promises to surpass that effort.

Besides neuro imaging, I also continue to conduct research in nonparametric statistics. A technique I have dubbed far casting cross validation (FCCV) improves ordinary cross validation in the face of serial/spatial correlation in the nonparametric regression setting for bandwidth selection. I am researching ways to improve FCCV working with a data driven continuous weighting mechanism. I have also developed multivariate response versions of univariate recursive partitioning classification methods and continue developing source code for a library of related functions.
 
 RESEARCH INTERESTS
 
Multivariate nonparametric function estimation
Statistical computing
Discrimination and classification
Multivariate statistics/data mining
Spatial modeling
 
 RECENT PUBLICATIONS
 
Carmack, P.S., Schucany, W.R., Spence, J.S., Gunst, R.F., Lin, Q., Haley, R.W., "Far Casting Cross Validation" Journal of Computational and Graphical Statistics, To appear. 2008
Haley, R.W., Spence, J.S., Carmack, P.S., Gunst, R.F., Schucany, W.R., Petty, F., Devous, Sr., M.D., Bonte, F.J., Trivedi, M.H., "Abnormal Brain Response to Cholinergic Challenge in Chronic Encephalopathy from the 1991 Gulf War" Psychiatry Research: Nueroimaging, To appear. 2008
Spence, J., Carmack, P.S., Gunst, R.F., Schucany, W.R., Woodward, W.A., and Haley,, "Accounting for Spatial Dependence in the Analysis of SPECT Brain Imaging" Journal of the American Statistical Association, Vol. 102, No. 478:464-473, 2007
Spence, J., Carmack, P.S., Gunst, R.F., Schucany, W.R., Woodward, W.A., and Haley,, "Using a White Matter Reference to Remove the Dependency of Global Signal" NeroImage, 32:49-53, 2006
Carmack PS, Spence J, Gunst RF, Schucany WR, Woodward WA and Haley RW, "Improved agreement between Talairach and MNI coordinate spaces in deep brain regions" NeuroImage, 22:367-71, 2004
 
 
Point and right click (click and hold for Mac users) your mouse on and select "Save this link (or target) as..." option to save the file to your local computer.