Non-rigid Medical Image Registration Using Image Field in Demons Algorithm

Sheng Lan, Zhenhua Guo, Jia You

Research output: Journal article publicationJournal articleAcademic researchpeer-review

Abstract

In medical imaging application, non-rigid registration is an important step that requires both accuracy and efficiency to align the deformed query image with the reference image. For non-rigid registration, Demons is generally used as an effective algorithm. However, since the computational model of Demons uses only the gradient information, the directional information of the image has not been fully utilized. In this paper, we proposed a novel non-rigid registration method for processing medical images by introducing the image field to the Demons algorithm. As direction information is very important for spatial transformation in registration and the image fields contain the direction of the image, by combing image fields with the traditional model based registration algorithm, we introduce the orientation field directly to the model for estimating the distortion, and thus make a better use of the direction information of images and simplify the deformation model. Different performance measures are used to evaluate the qualitative measures of the proposed scheme. Experiments on brain images and fundus images show the improvement of the proposed algorithm compared to the other existing state-of-the-art methods.
Original languageEnglish
Pages (from-to)48-57
JournalPattern Recognition Letters
Volume125
DOIs
Publication statusPublished - Jul 2019

Cite this