PET/SPECT/CT Image Processing, Image Reconstruction, 3D Geometry, Motion Estimation and Correction, Biomedical Signal Processing, Multi-resolution filtering, Learning algorithms (Support Vector, Neural networks)
Ph.D. in Electrical & Computer Engineering, University of Houston, Houston,
of an EEG-based automated neonatal seizure detection system
B.E. in Electrical
Engineering, Jadavpur University,
Kolkata, India 1997-2001
Department of Radiology, University
of Massachusetts Medical School, Worcester, MA 2007-Present
Instructor with Dr. Michael King 2009-Present
Research Fellow with
Dr. Michael King 2007-2009
- Sparse Reconstruction (Ongoing) - Investigation of novel reconstruction methods for low-count emission imaging, and low-cost multi-pinhole SPECT systems.
- Data-driven Motion Estimation in SPECT (Ongoing) – Patient motion
estimation from SPECT projection data with little or no information from additional surrogate motion signals. Strategies investigated are 2D-3D Registration and combined optimization of motion and activity using maximum likelihood.
- Analytical and Iterative Fan-beam algorithms for SPECT – Investigated
different strategies for attenuation correction in analytical reconstruction. Implemented a novel shift-variant resolution compensation method for better
image quality within iterative reconstruction, and conducted
an image quality comparison study at multiple count levels between Parallel
and Fan-beam SPECT for brain perfusion imaging using Tc99m (HMPAO).
- Motion estimation - Separation of respiratory motion and patient body motion
from a signal acquired using a motion-tracking system for correcting SPECT images.
Methods include Total-variation based smoothing, SVD & Quaternion-based
algorithms for estimation of motion.
Electrical and Computer Engineering
Department, University of Houston, Houston, TX 2002-2006
Graduate Research Assistant with Dr. John Glover
- Automated Neonatal Seizure Detection system using EEG signal. Methods
include signal analysis using Wavelets, Neural Networks, Clustering
- Implemented algorithms for application of Total Variation-based
denoising to Ultrasound Images. Studied the de-noising performance under
different noise models. (Course Project)
- Implemented different neural network architectures (Back Propagation
Algorithms, Error-based and Entropy-based minimization routines)
Indian Statistical Institute, Kolkata, India
Summer Intern with Dr. Sarbani Palit, Dr. Debashis Sengupta 2001
Analysis of Functional MRI data by Wavelet Decomposition. Work involved reconstruction of 3-dimensional activation produced in the brain by finger-tapping using wavelets.
AWARDS AND HONORS
- IEEE Senior Membership in 2011.
- Trainee Grant by IEEE Medical Imaging Conference for presenting work at the conference in 2007, 2008 and 2009. Grant is awarded to selected papers from the applicant pool by peer-reviewed scoring of abstracts.
- A co-authored paper presented at Society of Nuclear Medicine Conference, New Orleans, LA, 2008 (Conference pub. 19) was awarded best paper.
- Travel Fellowship awarded by American Clinical Neurophysiology Society to attend Annual meeting in 2006.
- Cullen & Ehrhardt fellowship by the Department of Electrical & Computer Engineering at the University of Houston in 2004-2006.
Title : Image reconstruction including shift-variant blur compensation, filed by Philips Medical Systems, CA, in 2009.
Link : http://patentscope.wipo.int/search/en/WO2011007270
Languages: MATLAB (Advanced), Linux Scripting in Bash shell, C (Advanced), C#(beginner),C++,
ITK, FLTK, VTK (as part of course work), assembly language programming (Motorola 68HC11 in
struction set on HandyBoard), Statistical Analysis using SAS
Platforms: Windows, Linux
Other: Experienced with NCAT phantoms, Monte-Carlo simulation of SPECT geometries using SIMIND