"Anatomic Morphologic Analysis of MR Brain Images"

Collaborating Institutions:

Massachusetts General Hospital Harvard Medical School
Draper Laboratory
Northeastern University
Boston University, Multi-Dimensional Signal Processing Lab
Massachusetts Institute of Technology

Key Personnel:

Abstract

Changes in the anatomical structure of the human brain occur as a normal part of the development and aging process, and in response to many disease and injury processes. In addition, recent success in functional neuroimaging has resulted in significant interest in the relationship between structural morphology and function. The ability to detect and monitor structural changes directly influences our ability to diagnose, treat and understand changes which associated with deleterious conditions. Magnetic Resonance Imaging (MRI) has been shown to be the most effective high-resolution, noninvasive means of assessing the structure of the intact human brain. Functional measures, such as metabolism and cerebral vascular hemodynamics, are important adjuncts to the diagnostic process. The goal of this grant is the quantitation of significant features in MRI images of the brain and their temporal evolution in a robust and reproducible fashion. To achieve this objective, we propose to improve quantitation of morphological features of brain structures through (i) improved automation of the segmentation and classification task, and (ii) improved representation and quantitation of the shapes of such structures. Several unique aspects of this proposal distinguishes it from past work performed in this area. First, a unified framework for segmentation and classification, as well as quantitation and representation of morphology in three dimensions is suggested. Second, this unified framework will incorporate the multi-spectral nature of MRI data, and will yield characterizations of uncertainties associated with the estimation process - thus permitting the rational analysis of the sensitivity of the techniques developed. Third, the approach set forth a systematic and incremental means of evaluating pixel classification performance - from phantoms to research quality to clinical quality data. Fourth, the approach expands upon traditional "static" image analysis by incorporating and extracting time evolution statistics for data analysis and anomaly screening.

Specific Aims

Aim 1. Build upon our initial progress in the development of a comprehensive methodology for effecting pixel classification and morphological quantification in single, vectored, and time-series MR images.

1.1 Integrate and test the efficacy of using our newly developed pixel classification methods which use variational calculus-based multi-spectral analysis and statistical modeling in an existing user-interface for routine morphometric analysis.

1.2 Extend the analysis methods to classification of tensor-valued diffusion-weighted MRI data and time-domain functional MRI data and continue development of a unified segmentation framework. Specifically, the simultaneous smoothing and segmentation formulation devised thus far can be used to smooth these highly noisy data sets, while retaining the sharp discontinuities between anisotropic regions to perform identification and mensuration of structurally and functionally defined regions.

1.3 Optimize the classification methods for specific clinical research applications. In particular, we will streamline the analysis method to provide rapid volumetric analysis of caudate and putamen in Huntington's disease and lesion localization in stroke patients.

Aim 2. Continue development of techniques for characterizing shape and change and explore methods which identify suitable shape characterization features.

2.1 Develop methods for efficient and robust characterization of two- and three-dimensional shapes in MR imagery, based upon extensions of our variational shape skeletonization technique. Techniques for characterizing shape evolution and anomalies will also be developed.

2.2 Explore novel and efficient adaptive shape representation techniques, and assess their utility in characterizing shape evolution.

2.3 Apply these shape and shape change characterization methods to specific clinical application areas. In particular, we will characterize caudate shape in the normal population and contrast this to caudate shape as observed in patients with Huntington's disease. Applications with respect to monitoring the progression of brain tumors will also be examined.

Aim 3. Continue the dissemination of information to the Neuroscience community through the expansion the "Internet Brain Segmentation Repository" (IBSR), the development of a "Internet Brain Volumetric Database" (IBVD), and the continued distribution of software developed at our site.

3.1 Establish a web-based automated clearing house for Human Brain Project related information including, but not limited to, grant recipients and their web sites, related meetings, talks and presentations supported, or related to HBP activities, published papers, etc. This will be designed to be self-sustaining, so that individuals submit and maintain the information which is documented at this site. This will serve as a self-documenting archive of Human Brain Project activities.

3.2 Facilitate the dissemination of volumetric data. This will include dissemination of group and individual volumetric data generated by the Center for Morphometric Analysis, as well as an initial feasibility study of the design, creation and initialization of an IBVD. Specifically, this site will contain a web-based user interface for data exploration and entry, as well as a database of individual and group volumetric data and subject characterizations.

3.3 Maintain and support information dissemination through the Project Web Site, including software distribution, access to the IBSR, as well as to other related Internet sites.

Publications resulting from this research:

1. JF. Bates, JW. Meyer, N. Makris, DN. Kennedy, JW. Belliveau and VS. Caviness Jr. "Parcellation of Subcortical Gray Matter Structures in the Human Brain: A Morphometric Tool with Functional Applications". Neuroimage:3(3), p S127, 1996.

2. A. Jiang, DN. Kennedy, JR. Baker, RR. Benson, BR. Rosen and JW. Belliveau. "Morphologic Region Analysis of Statistical Parameter Maps in fMRI". Neuroimage:3(3), p S67, 1996.

3. N. Makris, JD. Schmahmann, DN. Kennedy, RR. Benson and VS. Caviness Jr. "Human Cerebellar Cortex: MRI-Based Topographic Parcellation for Localization and Morphometry". Neuroimage:3(3), p S140, 1996.

4. N. Makris, DN. Kennedy, AJ. Worth, G. Sorensen, GM. Papadimitriou, TG. Reese, BR. Rosen, VS. Caviness Jr., DN. Pandya and E. Kaplan. "Conduction Aphasia: In vivo Demonstration with Diffusion Weighted Imaging (DWI)". Neuroimage:5(4), p S573, 1997.

5. N. Makris, AJ. Worth, G. Sorensen, GM. Papadimitriou, O. Wu, TG. Reese, VJ. Wedeen, TL. Davis, VS. Caviness Jr., BR. Rosen, DN. Pandya and DN. Kennedy. "In vivo topographic characterization of cortico-cortical association pathways in the human with diffusion weighted imaging (DWI)," Neuroimage:5(4), p S430, 1997.

6. JW. Meyer, N. Makris, JF. Bates, DN. Kennedy and VS. Caviness Jr. "Computational Parcellation of Cerebral White Matter in the Human Brain". Neuroimage:5(4), p S401, 1997.

7. D.N. Kennedy, A.J. Worth, V.S. Caviness, Jr., "MRI-Based Internet Brain Segmentation Repository," Proc. International Society Magnetic Resonance in Medicine, Vol. 4(3), p 1657, 1996.

8. JD. Schmahmann, J. Doyon, C. Holmes, N. Makris, M. Petrides, DN. Kennedy and AC. Evans. "An MRI Atlas of the Human Cerebellum in Talairach Space," Neuroimage:3(3), p S122, 1996.

Manuscripts: (Published, in press, or submitted)

1. HC. Breiter, RL. Gollub, RM. Weisskoff, DN. Kennedy, N. Makris, JD. Berke, JM. Goodman, HL. Kantor, DL. Gastfriend, JP. Riorden, RT. Mathew, BR. Rosen and SE. Hyman. "Acute Effects of Cocaine on Human Brain Activity and Emotion", Neuron:19, p591-611, 1997.

2. N. Makris, AJ. Worth, G. Sorensen, GM. Papadimitriou, O. Wu, TG. Reese, VJ. Wedeen, TL. Davis, JW. Stakes, VS. Caviness Jr., E. Kaplan, BR. Rosen, DN. Pandya and DN. Kennedy, "Morphometry of In Vivo Human White Matter Association Pathways with Diffusion-Weighted Magnetic Resonance Imaging". Ann Neurol:42, In Press, 1997.

3. J. Kaufhold, M. K. Schneider, W. C. Karl, and A. S. Willsky, "A Recursive Estimation Approach to the Segmentation of MR Imagery," Intl J Patt Recog and Artif Intell, In Press, 1997.

4. J. Kaufhold and W. C. Karl, "A Recursive Estimation Approach to the Segmentation of MR Imagery, submitted to International Conference on Image Processing," Santa Barbara, CA, October 26-29, 1996.

5. J. Kaufhold, M. K. Schneider, W. C. Karl, and A. S. Willsky, "A Recursive Estimation Approach to the Segmentation of MR Imagery," Workshop on MR Signal Processing, Univ. of Illinois at Urbana-Champaign, Urbana, Illinois, October 18-20, 1997.

6. DN. Kennedy, N. Lange, N. Makris, JW. Meyer and V.S. Caviness Jr. "Gyri of the Human Neocortex: An MRI-Based Analysis of Volume and Variance". Submitted.

7. D.N. Kennedy, N. Makris, J. F. Bates, V. S. Caviness Jr., "Structural Morphometry in the Developing Brain," in Developmental Neuroimaging, R.W. Thatcher, et al., Editors., Academic Press: San Diego., p. 29-41, 1996.

8. H. Pien, M. Desai, and J. Shah, "Segmentation of MR images using curve evolution and prior information," Int'l J. Pattern Recognition and Artificial Intelligence, In Press, 1997.

9. H. Pien, C. Karl, D. Kennedy, A. Worth, A. Willsky, (Editorial). International Journal of Pattern Recognition and Artificial Intelligence, In Press, 1997

10. M. Saeed, "Maximum Likelihood Parameter Estimation of Mixture Models and Its Application to Image Segmentation and Restoration," Masters Thesis, Department of Electrical Engineering and Computer Science, MIT, June, 1997.

11. M. K. Schneider, P. W. Fieguth, W. C. Karl, and A. S. Willsky, "Multiscale Statistical Methods for the Segmentation of Images," IEEE Trans Image Proc, Submitted, 1997.

12. M. Schneider, "Multiscale Methods for the Segmentation of Images," Masters Thesis, Department of Electrical Engineering and Computer Science, MIT, June, 1996.

13. M. Schneider, P. Fieguth, W.C. Karl, A.S. Willsky, "Multiscale Methods for the Segmentation of Images," Proc. Intl. Conf. Acoustics, Speech, and Signal Processing, Atlanta, GA, May 7-10, 1996,

14. J. Shah, "Shape recovery from noisy images by curve evolution," IASTED Int'l Conf. Signal and Image Processing, Nov 1995.

15. J. Shah, "Curve evolution and segmentation functionals: Applications to color images," IEEE Int'l Conf. Image Proc., 9, 461-464, 1996.

16. J. Shah, H. Pien, and J. Gauch, "Recovery of surfaces with discontinuities by fusing shading and range data within a variational framework," IEEE Trans Image Processing, 5(8), p1243-1251, 1996.

17. J. Shah, "A common framework for curve evolution, segmentation, and anisotropic diffusion," IEEE Conf. Computer Vision and Pattern Recognition, 6, 136-142, 1996.

18. S. Tari and J. Shah, "Local Symmetries of Shapes in Arbitrary Dimension," International Conference on Computer Vision, January, In Press, 1998.

19. S. Tari and J. Shah, "Simultaneous Segmentation of Images and Shapes," SPIE Conference on Vision Geometry, August, 1997.

20. S. Tari, J. Shah, and H. Pien, "A computationally efficient shape analysis via level sets," IEEE Wkshp. Mathematical Methods in Biomedical Image Analysis, 6, 1996.

21. S. Tari, J. Shah, and H. Pien, "Extraction of shape skeletons from grayscale images," J of Comp Vis and Image Understanding, 66(2), 133-146, 1997.

22. A.J Worth, Makris, N., Caviness, V. S. Jr., Kennedy, D. N, "Neuroanatomical Segmentation in MRI: Technological Objectives," Intl. J. Pat. Recog. Art. Intel., 1997. In Press.

23. A.J. Worth, Makris, N., Meyer, J. W., Caviness, V. S. Jr., Kennedy, D. N., "Automated Segmentation of Brain Exterior in MR Images Driven by Empirical Procedures and Anatomical Knowledge," in Proc Information Processing in Medical Imaging, Poultney, Vermont, June, Vol. 1230, pp. 99-112, 1997.

24. A.J. Worth, Makris N, Patti MR, Goodman JM, Hoge EA, Caviness VS, Jr., and Kennedy DN, "Precise Segmentation of the Lateral Ventricles and Caudate Nucleus in MR Brain Images using Anatomically Driven Histograms," Submitted, 1997.

25. A.J. Worth, Makris N, Meyer JW, Caviness VS, Jr., and Kennedy DN, "Semi-Automatic segmentation of brain exterior in MR images driven by empirical procedures and anatomical knowledge," Submitted, 1997.

Special Issues:

1. H. Pien, C. Karl, D. Kennedy, A. Worth, and A. Willsky, Editors, Special issue on Processing of MR Images of the Human Brain, in International Journal of Pattern Recognition and Artificial Intelligence, In Press, 1997.

2. J. Belliveau, P. Fox, D. Kennedy, B. Rosen and L. Ungerleider, Supplement Editors, Special Issue for Second International Conference on Functional Mapping of the Human Brain, in NeuroImage, 3(3), 1996.