1. Parcellation of Human Inferior Parietal Lobule Based on Diffusion MRI Bilge Soran 1 Zhiyong Xie 2 Rosalia Tungaraza 3 Su-In Lee 1 Linda Shapiro 1,2 Thomas Grabowski 3 University of Washington 1 Dept. of Computer Science and Engineering 2 Dept. of Electrical Engineering 3 Integrated Brain Imaging Center Aug 2012 This work was supported by NIH-NINDS Grant No. RC4-NS073008 (PI: T. Grabowski).

2. Motivation The function of a brain can be studied by analyzing its anatomical connections. The importance of this work is in its investigation of methods for parcellation of the brain of a living subject, rather than manual parcellation of a post- mortem subject. 2

3. Outline Background – Human inferior parietal lobule – Diffusion MRI Current Approaches Methods – Unsupervised Clustering Approaches Evaluation Metric Results 3

4. Human Inferior Parietal Lobule The IPL is the cortical region with marked functional heterogeneity involved in visuospatial attention, memory, and mathematical cognition functions. 4

5. Aim To parcellate the IPL data of living subjects and evaluate the parcellation quality by means of overlap with clusters of a standard atlas. 5 Juelich atlas A parcellation result in sagittal view

6. Diffusion MRI Measures the diffusion of water molecules in biological tissues. 6 Tractography Method for identifying anatomical connections in the living human brain. Offers an overall view of brain anatomy, including the degree of connectivity between different regions of the brain.

7. Current Parcellation Approaches 1.Methods relying on the known connection patterns of the functional fields. 2.Methods using a statistical model, based on the assumption of the statistical distribution of the data. 3.Methods using unsupervised machine learning methods. 7 In this project we used unsupervised machine learning techniques to parcellate the IPL region.

8. Our Method 8 Preprocessing Generate the connectivity map for each seed point using Probabilistic Tractography Cluster the voxels of the IPL into functional fields based on the connectivity pattern Using K-Means, EM, Spectral Clustering Verification with the Juelich atlas Using proposed metrics

9. Juelich Atlas (Juelich histological (cyto- and myelo-architectonic) atlas) 9 The Juelich atlas is a probabilistic map of the areas of the brain. It was created based on the microscopic and quantitative histological examination of ten human post-mortem brains. lh-IPC, Sagittal View 1 2 3 4 5 6 7

10. Preprocessing Regions of interests (ROIs) were extracted. For each voxel in the IPL, a connectivity profile is calculated using probabilistic tractography. 10  5000 particles initiated from each seed voxel  # of particles that reached each target region counted. Destrieux atlas 1 IPL region in Destrieux atlas 1 1 Destrieux C., et al. Automatic parcellation of human cortical gyri and sulci using standard anatomical nomenclature, NeuroImage, 2010

11. Data for Clustering 3D coordinates of the IPL voxels. Target connectivity probabilities of each seed voxel.  Target regions showing no connectivity patterns with any of the seed voxels are discarded. 11

12. Parcellation Methods K-means Clustering Mean Shift Clustering Expectation Maximization (EM) Clustering Spectral Clustering – Standard Normalized Graph Cuts – Normalized Graph Cuts with Feature Selection – Normalized Graph Cuts with K-means 12

13. Similarity Matrix Computation 13 Seed Voxels Target Regions Connectivity Matrix Seed Voxels Distance Matrix Seed Voxels Connectivity Similarity Matrix Seed Voxels Spatial Affinity Matrix Seed Voxels Composite Similarity Matrix +

14. Similarity Computation 14

15. Feature Selection Target regions having similar connectivity patterns for all seed voxels were detected and removed by looking at the variance of the columns of the connectivity matrix and applying a threshold. 15 Target Regions Seed Voxels

16. Evaluation Metric Based on the number of intersecting voxels between a cluster of the atlas and a normalized cut cluster. Defined as the average of the sum of the largest overlap with the computed clusters of each atlas cluster and the sum of the largest overlap with the atlas clusters of each computed cluster. 16 AC 1AC 2AC 3AC 4AC 5AC 6AC 7 CC 1 000911530 CC 2 000240798 CC 3 053679000 CC 4 678144002 CC 5 913100000 An example table used in evaluation.

17. Evaluation Metric 17 Else the resulting parcellation is unacceptable.

18. Experiments and Results The data set consisted of the left and right IPL regions, which have similar connectivity patterns, of 19 subjects. The right IPL data set was divided into two with 10 subjects for training and 9 for testing; the left IPL data set was used only for testing. This resulted in 10 subjects for training and 28 subjects for testing. 18

19. Experiments and Results The three different normalized cut algorithms were run on the training set, and the best parameter set for each algorithm was determined according to the proposed metric. Because the Juelich atlas contains 7 manually determined clusters, experiments were run with +/- 2 margin of 7, namely 5 to 9 clusters. Experimented with α varying from 0.5 to 0.9 with 0.1 intervals. 19

20. Normalized Graph Cuts After the best parameters were determined on the training set, both NGC and NGC with feature selection algorithms were run on the whole training and testing sets. The normalized graph cuts with the K­-means approach did not produce any acceptable parcellation results according to the proposed metric. 20

21. SubjectRight HemisphereLeft Hemisphere NGC-stdNGC-feaNGC-stdNGC-fea 1 0.6253920.6269590.6154370.620902 2 0.7535770.7519870.6090910.610909 3 0.7184870.7100840.6041260.608055 4 0.5699520.5707470.6726940.674503 5 0.6520210.6555360.6756520.672174 6 0.6378830.6462400.7063880.707617 7 0.640599 0.5537920.554674 8 0.5934150.6010720.5926570.599650 9 0.6276220.6284970.6364940.635057 10 0.6832120.6824820.5668360.564298 11 0.590669 0.6275590.633071 12 0.6114520.6155420.6406250.637336 13 0.7479170.7468750.6209480.622195 14 0.6577250.6534330.6913210.683432 15 0.666667 0.592619 16 0.6575880.7013620.6604170.658333 17 0.6007680.5998080.642747 18 0.611494 0.7169600.712555 19 0.6708020.6698470.6444310.643819 AVERAGE0.6482760.6510470.6353050.635471 Parcellation performances 21

22. Juelich atlas and parcellation results in sagittal view 22

23. Conclusions This work investigates methods for parcellating the IPL region of a living subject. We used a new atlas-based metric to evaluate the quality of the parcellation. Normalized graph cuts showed the best performance among the clustering methods applied. Our work shows the feasibility of this approach for parcellation of brain regions of living subjects. 23

24. References [1] P. G. P. Nucifora, et al. Diffusion-Tensor MR Imaging and Tractography: Exploring Brain Microstructure and Connectivity. Radiology, 245(2):367–384, 2007. [2] S. Jbabdi and M. W. Woolrich and T. E. Behrens. Multiple-subjects connectivity-based parcellation using hierarchical Dirichlet process mixture models. NeuroImage, 44(2):373–384, 2009. [3] A. Anwander, et al. Connectivity-Based Parcellation of Broca’s Area. Cereb. Cortex, 17(4):816–825, 2007. [4] R. B. Mars, et al. Diffusion-Weighted Imaging Tractography-Based Parcellation of the Human Parietal Cortex and Comparison with Human and Macaque Resting-State Functional Connectivity. The Journal of Neuroscience, 2011. [5] M. Ruschel, et al. Connectivity-based parcellation of the human inferior parietal cortex - New insights into the problem of structurefunction relationships in humans and macaques, 2009. [6] B. Fischl, M. I. Sereno, and A. Dale. Cortical surface-based analysis: II: Inflation, flattening, and a surface- based coordinate system. NeuroImage, 9(2):195 – 207, 1999. [7] J-D. Tournier, F. Calamante, and A. Connelly. Robust determination of the fibre orientation distribution in diffusion mri: Non-negativity constrained super-resolved spherical deconvolution. NeuroImage, 35(4):1459 – 1472, 2007. [8] MathWorks. http://www.mathworks.com/help/toolbox/stats/pdist.html, 2011. [9] J. Shi and J. Malik. Normalized Cuts and Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(8):888–905, 2000. [10] U. von Luxburg. A tutorial on spectral clustering. Technical Report 149, Max Planck Institute for Biological Cybernetics, 2006. [11] S. Caspers, et al. Probabilistic fibre tract analysis of cytoarchitectonically defined human inferior parietal lobule areas reveals similarities to macaques. NeuroImage, 2011. 24