Suyash P. Awate Suyash P. Awate
Associate Professor
Computer Science and Engineering Department, Indian Institute of Technology (IIT) Bombay
Office: A-214, Kanwal Rekhi Building
Email: my-first-name @cse.iitb.ac.in
   
Research Publications Teaching Students CV Personal
 
Segmenting SPD-Tensor-Valued Images in Diffusion MRI
 
Suyash P. Awate, Hui Zhang , James C. Gee
A Fuzzy, Nonparametric Segmentation Framework for DTI and MRI Analysis: With Applications to DTI-Tract Extraction
IEEE Trans. Med. Imaging 2007, 26(11):1525-1536

Suyash P. Awate, James C. Gee
A Fuzzy, Nonparametric Segmentation Framework for DTI and MRI Analysis
Proc. Information Processing in Medical Imaging 2007, Springer LNCS 4584 pp. 296-307
 
Challenges using Streamline Tractography to Extract Cingulum

dashed-line = ground truth
(1) streamlines terminate before the bend

(2) streamlines leak out of the cingulum
 
Modeling Statistics of Tensors in Fiber Bundles

Tensors in fiber bundles can lie on arbitrary hyper-surfaces in the Riemannian space of diffusion tensors (i.e. SPD matrices).

We infer the probability density function of the tensors via kernel density estimation in the Riemannian space.

The density estimates can be shown to converge by combining results from the (i) Log-Euclidean framework [Arsigny et al.] and (ii) kernel density estimation [Pelletier et al.]
 
Segmentation Results

Initialize via (i) probabilistic atlas (3D) or (ii) manually (one slice suffices)


 
Comparing Tractography and Segmentation

(1) streamline tractography


(2) segmentation
 
Related Works
 
Hui Zhang, Suyash P. Awate, Sandhitsu R. Das, John H. Woo, Elias R. Melhem, James C. Gee, Paul A. Yushkevich
A tract-specific framework for white matter morphometry combining macroscopic and microscopic tract features
Medical Image Analysis 14(5): 666-673 (2010)

Lenglet C, Campbell JS, Descoteaux M, Haro G, Savadjiev P, Wassermann D, Anwander A, Deriche R, Pike GB, Sapiro G, Siddiqi K, Thompson PM.
Mathematical methods for diffusion MRI processing.
NeuroImage 2008

A Goh, C Lenglet, PM Thompson, R Vidal
A nonparametric Riemannian framework for processing high angular resolution diffusion images and its applications to ODF-based morphometry.
NeuroImage 2011

Marc Niethammer, Christopher Zach, John Melonakos, Allen Tannenbaum
Near-tubular fiber bundle segmentation for diffusion weighted imaging: Segmentation through frame reorientation
NeuroImage 2009

Z. Wang, B. Vemuri
DTI segmentation using an information-theoretic tensor dissimilarity measure
IEEE Trans. Medical Imaging 2005, 24(10):1267-1277

B. Pelletier
Kernel density estimation on Riemannian manifolds
Stat. and Prob. Letters 2005, 73:297-304

V. Arsigny, P. Fillard, X. Pennec, N. Ayache
Geometric means in a novel vector space structure on symmetric positive-definite matrices
SIAM J. Matrix Analysis and Applications 2007, 29(1):328-347