On Two Algorithms for Multi-Modality Image Registration Based on Gaussian Curvature and Application to Medical Images

Nasra Begum,Muniba Ashfaq,Hadia Atta,Nasru Minallah,Noor Badshah,Mazlinda Ibrahim

doi:10.1109/access.2021.3050651

Abstract

Registration of multi-modal images is one of the challenging problems in image processing nowadays. In this paper, two novel non-rigid registration models are proposed for multi-modality images. In model 1, mutual information of the template and reference images is used as data fitting term with Gaussian curvature regularization. This approach may not give satisfactory results in noisy images or images having bias field. To overcome this drawback, model 2 is proposed which is based on normalized gradient of both template and reference images as a data fitting term instead of mutual information. To get best transformations, both the models are minimized by using Augmented Lagrangian Method. The proposed models can register multi-modality images without effecting edges and other important fine details and are also tested on various medical images like (T1-T2 MRI, PD weighted-T2 MRI) noisy and synthetic images. The proposed models are also tested on a well known free available Brainweb dataset, where they produced satisfactory results. From experimental results, it can be observed that normalized gradient field based model gives better results than mutual information based model. Comparison is done qualitatively and quantitatively through Jaccard Similarity Coefficient.

Highlights

Image registration is one of the most important and challenging task in medical imaging, which aims on finding an optimal transformation for alignment of different images data
We propose the following functionals for minimization to register two multi-modality images: Energy Functional 1: The energy functional for our first new proposed model is based on mutual information (MI)
EXPERIMENTAL RESULTS the performance of the proposed models is assessed by using numerical experiments to examine the robustness and efficiency of the algorithm for multi-modality images

Summary

Introduction

Image registration is one of the most important and challenging task in medical imaging, which aims on finding an optimal transformation for alignment of different images data. Image registration is widely used in art, astronomy, criminology, cartography, computer vision, biological imaging, remote sensing and especially in medical imaging for diagnosis,monitoring of tumor growth and for therapy guidance [1]–[5]. General Frame Work: For given template image T and reference image R, defined on ⊆ Rd , d ∈ N is the dimensionality of the data with smooth boundary ∂. The basic idea of image registration is to find a transformation (u)(·) : → such that (u(x)) = x + u(x). It is enough to find a deformation or displacement field u : → such that the transformed template image

Objectives

Methods

Results

Conclusion