Registration of MRI-CT Images of the Human Brain using Deep Learning

Please enable javascript in your browser.

Ansarino, Keyvan | 2023

66 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 56203 (05)
University: Sharif University of Technology
Department: Electrical Engineering
Advisor(s): Fatemizadeh, Emadeddin
Abstract:
Image registration is the process of matching the coordinate systems of two or more images. Medical image registration has been used in a variety of applications such as segmentation, motion tracking and etc. Recently, the use of deep neural networks has been demonstrated as a useful approach to registration problems. In this work, we propose two separate novel Convolutional Neural Network (CNN) architectures for multi-modal rigid and affine registration of the CT-MRI images of the brain. A dataset consisting of CT-MRI images of 37 subjects was used for training and evaluation of the networks. For both networks, the proposed models achieved high mutual information value between predicted CT images and their corresponding MRIs (0.657 for rigid, and 0.633 for affine) and a mean dice score of 0.984 for rigid registration. The above-mentioned results are very close to the values achieved from ground-truth images, which were already existing in the used dataset. So, it can be said that our designed networks have had ideal performances in rigid and affine registration of two-dimensional CT-MRI images of the brain
Keywords:
Image Registration ; Convolutional Neural Network ; Rigid Registration ; Deep Learning ; Magnetic Resonance Imagin (MRI) ; Affine Registration ; CT Scan

No TOC