All research projects in the program focus on image segmentation. Evaluating performance of computer-based segmentation methods requires considering both geometric and medical measures. Human and machine observer studies provide a necessary and compelling basis for such evaluation. Consequently a number of such studies are included in this program project. The core personnel will lead in study design, conduct the required statistical power and data analyses, and manage the associated data. The research program will be improved directly by optimal design and analysis, and indirectly by the scientist time freed. We shall also support the program by adapting recently developed statistical methods which allow more efficient designs than those presently available. Nearly all planned analyses involve the use of repeated measures Analysis of Variance (ANOVA). All research also uses a multiple study strategy. Statistical power analysis provides a key tool in increasing efficiency. Recent advances in """"""""internal pilot designs"""""""" for univariate linear models data and power analysis will be adapted to the research in this project. Subsequent publications will also help disseminate the new methods to the medical imaging community. The core activity has four specific aims: (1) Design. Implement the principles of design. (2) Analysis. Conduct all statistical data analyses for project studies. (3) Research Data Management. Archive and manage all data resulting from statistical analysis and study planning for project studies. (4) Adapt Methods. Choose designs, power and data analyses which allow using recently developed methods for """"""""internal pilot studies.""""""""
| Sadeghi, Neda; Prastawa, Marcel; Fletcher, P Thomas et al. (2012) STATISTICAL GROWTH MODELING OF LONGITUDINAL DT-MRI FOR REGIONAL CHARACTERIZATION OF EARLY BRAIN DEVELOPMENT. Proc IEEE Int Symp Biomed Imaging :1507-1510 |
| El-Sayed, Mohamed; Steen, R Grant; Poe, Michele D et al. (2010) Brain volumes in psychotic youth with schizophrenia and mood disorders. J Psychiatry Neurosci 35:229-36 |
| Chi, Yueh-Yun; Muller, Keith E (2010) Using scientifically and statistically sufficient statistics in comparing image segmentations. Stat Interface 3:91-101 |
| Simpson, Sean L; Edwards, Lloyd J; Muller, Keith E et al. (2010) A linear exponent AR(1) family of correlation structures. Stat Med 29:1825-38 |
| Clement-Spychala, Meagan E; Couper, David; Zhu, Hongtu et al. (2010) Approximating the Geisser-Greenhouse sphericity estimator and its applications to diffusion tensor imaging. Stat Interface 3:81-90 |
| Edwards, Lloyd J; Muller, Keith E; Wolfinger, Russell D et al. (2008) An R2 statistic for fixed effects in the linear mixed model. Stat Med 27:6137-57 |
| Hemminger, Bradley M; Bauers, Anne; Yang, Jian (2008) Comparison of navigation techniques for large digital images. J Digit Imaging 21 Suppl 1:S13-38 |
| Crouch, Jessica R; Pizer, Stephen M; Chaney, Edward L et al. (2007) Automated finite-element analysis for deformable registration of prostate images. IEEE Trans Med Imaging 26:1379-90 |
| Gurka, Matthew J; Coffey, Christopher S; Muller, Keith E (2007) Internal pilots for a class of linear mixed models with Gaussian and compound symmetric data. Stat Med 26:4083-99 |
| Muller, Keith E; Edwards, Lloyd J; Simpson, Sean L et al. (2007) Statistical tests with accurate size and power for balanced linear mixed models. Stat Med 26:3639-60 |
Showing the most recent 10 out of 69 publications
