Multi-modal Image Registration for Deep Brain Stimulation Analysis
Harikrishna G. N. Rai. Krishnamurty Sai Deepak and P. Radha Krishna
*General Electricals, Bangalore, India, E-mail:harikrishna.raign@gmail.com Infosys Limited, Bangalore, India, E-mail: krishnamurtysai d@infosys.com Department of Computer Science and Engineering, National Institute of Technology (NIT), Warangal,India, E-mail: prkrishna@nitw.ac.in
Received in final form on March 30, 2021
Abstract
Deep Brain stimulation (DBS) surgery is microelectrode-guided surgery
for the treatment of patients with movement disorders. DBS on deep
brain structures requires effective postoperative procedures for analysis
of surgical success. Thus, brain shift analysis is an important stage of
DBS surgery where pre-operative and postoperative images are compared
to measure the brain shift. Since information gained from two images
acquired in the clinical track of events is usually complementary, properly
integrating useful data obtained from the separate images is challenging.
The integration process to bring the modalities involve spatial alignment,
a procedure referred to as registration. Registration is typically followed
by a fusion step required for the integrated display of the image data.
This paper presents a novel approach that focuses on multimodality data
fusion to analyze the pre-operative and postoperative CT and MR images.
Our multimodal image registration model considers brain images based on
a projection-based reference image. The registration is based on sharing
mass in a predetermined direction: axial, sagittal, and coronal. Then,
the region of interest is registered through modified mutual information
based on the overlap rate and a weight assigned to it. Our approach is a
combination of both coarse and fine registration techniques.
Keywords
Deep brain stimulation, coarse and fine registration, CT and MR images.
Cite This Article
Harikrishna G. N. Rai, Krishnamurty Sai Deepak and P. Radha Krishna,Multi-modal Image Registration for Deep Brain
Stimulation Analysis, J. Innovation Sciences and Sustainable Technologies, 1(2)(2021), 189-204. https://doie.org/10.0608/JISST.2022729673
78 5 Download