Open Access

ARTICLE

Nonlinear Registration of Brain Magnetic Resonance Images with Cross Constraints of Intensity and Structure

Han Zhou^1,2, Hongtao Xu^1,2, Xinyue Chang^1,2, Wei Zhang^1,2, Heng Dong^1,2,*

1 Digital Fujian Research Institute of Big Data for Agriculture and Forestry, Fujian Agriculture and Forestry University, Fuzhou, 350002, China
2 College of Computer and Information Science, Fujian Agriculture and Forestry University, Fuzhou, 350002, China

* Corresponding Author: Heng Dong. Email:

Computers, Materials & Continua 2024, 79(2), 2295-2313. https://doi.org/10.32604/cmc.2024.047754

Received 16 November 2023; Accepted 04 March 2024; Issue published 15 May 2024

Abstract

Many deep learning-based registration methods rely on a single-stream encoder-decoder network for computing deformation fields between 3D volumes. However, these methods often lack constraint information and overlook semantic consistency, limiting their performance. To address these issues, we present a novel approach for medical image registration called the Dual-VoxelMorph, featuring a dual-channel cross-constraint network. This innovative network utilizes both intensity and segmentation images, which share identical semantic information and feature representations. Two encoder-decoder structures calculate deformation fields for intensity and segmentation images, as generated by the dual-channel cross-constraint network. This design facilitates bidirectional communication between grayscale and segmentation information, enabling the model to better learn the corresponding grayscale and segmentation details of the same anatomical structures. To ensure semantic and directional consistency, we introduce constraints and apply the cosine similarity function to enhance semantic consistency. Evaluation on four public datasets demonstrates superior performance compared to the baseline method, achieving Dice scores of 79.9%, 64.5%, 69.9%, and 63.5% for OASIS-1, OASIS-3, LPBA40, and ADNI, respectively.

---

Keywords

---