In the field of image processing research at home and abroad, a lot of image registration research methods and work have been reported, a large number of image registration ideas have been generated, and a variety of image registration algorithms and applications have been established, such as gray-scale information methods, including transform domain method and feature-based method, etc. Each of them derives a variety of specific methods to conduct multi-directional research on registration technology and matching accuracy. Various methods are oriented to a certain range of application fields with their own characteristics[1-3].

In recent years, with the rapid development of imaging technology, such as CT, EEG, MRI, DTI, SPECT and other technologies, a large number of different image data have been generated. Great changes have taken place in the spatial and temporal dimensions of the image. Processing has more stringent requirements in different granularities and different dimensions[4-6]. Of course, with the introduction of many new theories and methods in various disciplines, around image processing and technology, experts and scholars at home and abroad have proposed many matching technologies that are combined with specific theories, methods and tools, such as matching based on neural networks. Technology, matching technology based on wavelet analysis and transformation, etc. At the same time, it also provides a variety of solutions and platforms for different image processing needs, as well as the development and application of medical image algorithms[7].

In short, as a typical problem and technical difficulty in the field of image processing research, image registration will have more research, methods or platforms. However, each specific research, method and platform will be more specific, in-depth and detailed. Every stage, every granularity, every dimension and every precision of image processing will get higher and higher.

Object of the design

The main purpose of this research is to use the mapping algorithm of image space sequence to realize the image registration of 3D medical image images to standard space. Medical image expression is not only getting finer in granularity, higher and higher in time and space resolution, but also richer and richer in expression. Different image expression methods have enriched medical image processing methods. Medical image registration is one of the basic operations in medical research and clinical applications. The basic processing of image problems by using certain image expressions can achieve the purpose of image application, thereby providing for subsequent medical image processing, especially in the field of functional and structural images[6-9]. Lay the foundation for subsequent operations.

Method of the design

The method of this research is to transform the problem of image registration into the same spatial sequence expression and mapping to realize the function of image registration. The problem of image registration is how to compare images and how to generate image representation in standard space under the same template. This research is based on a unified spatial sequence expression for mapping under a standard space, which can achieve the requirements of image spatial registration, and realize the image registration by using the conversion and calculation of the mathematical expression of the spatial sequence[9-12].

This research builds on the advantages and foundation of image processing by Matlab, establishes a mapping method through different expressions of images, completes registration and verification through mutual mapping, and forms an operable software program to solve the problem of image registration.

Process of the design

First, image spatial sequence extraction: Obtain the spatial sequence expression of the image by obtaining various expressions of the original image, and obtain data such as the way and content of the spatial sequence expression extracted.

Second, expression of the standard space of the sphere: According to the requirements of the image space, select the definition, expression and parameters of the standard space of the sphere.

Third, the mapping of the image space sequence to the sphere: According to the actual spatial meaning of the image, a mapping algorithm to the standard space of the sphere is designed to obtain the image space sequence in the standard space. Furthermore, various calculations and expressions are performed on the image according to the definition of the sphere, so as to realize the registration of the image.

Forth, feedback from the sphere to the image: the image sequence in the standard space is fed back to the original image space sequence to verify the accuracy and effectiveness of the mapping algorithm for registration.

Fifth, Matlab image processing: programming to realize various calculations, expressions and algorithms in the process of designing image processing.

Image registration is a typical technical problem in the field of image processing research. In medical image processing, image registration is the basis and basis for image region identification and spatial calculation. It is also an important basis for supporting the correctness of results and conclusions, especially for accurate In terms of the research and application of identification and calculation, a reasonable and more accurate image registration method has an important interpretation and support for the results and significance of the research[3-6].

With the development of medical imaging technology, various imaging equipment generates a large number of different formats of medical imaging images, which have reached a subtle level of expression in time and space resolution, forming a large number of 3D space medical images. Due to the individual differences in images and the differences between individuals in image collection, it forms a vastly different image storage. In research and application, it is necessary to form a large number of images into a unified space and result representation under a template to continue subsequent processing to meet statistical requirements. Therefore, it is of great significance for medical image processing and application to register the difference images formed under different modalities into a unified standard space[4-8].Due to different acquisition methods and equipment, different storage formats, different representations, and different application environments, the image is converted into a unified spatial sequence, and correction is required to perform the mapping. At the same time, the selection of the mapping algorithm also affects the final image registration Therefore, spatial sequence mapping is the main problem to be solved[13].Image data processing will involve multiple related packages and source programs, especially this project needs to process the data from the bottom, so the program needs to be redesigned to realize the complete data processing process[14].

Starting from the time sequence expressed in the image space, the medical images in the 3D space are uniformly registered to the standard space through the registration conversion algorithm, and then the data in the standard space is fed back to the source image data to achieve image registration. purpose. Any images obtained on the same acquisition device and the same parameter settings have the same spatial sequence, and the spatial sequence is mapped to a unified spherical standard space to form an image in the standard space to realize the image registration function. The calculation and realization of the project are based on the Matlab platform as the experimental environment.

The spatial sequence of the original image and the mapping of the spherical space are used to realize the registration requirements of medical images, provide a basis for the subsequent processing of medical images, and provide method support for image registration, and provide support for more accurate research results.

Acknowledgements

This research was supported by the National Students’ project for innovation and entrepreneurship training program under Grant Number S201910439025

The authors are grateful to the anonymous referees for their valuable comments and suggestions.

The authors declare that they have no conflict of interest

No funding sources

The study was approved by the Shandong First Medical University and Shandong Academy of Medical Sciences, Taian, China.

1. Hill, Derek L. G., Philipp G. Batchelor, Mark Holden, et al. "Medical Image Registration." Physics in Medicine & Biology 31.4 (2008): pp. 1-45. https://doi.org/10.1088/0031-9155/31/4/001.

2. Aladl, U. E., and T. Peters. "Medical Image Registration." Multi Modality State-of-the-Art Medical Image Segmentation and Registration Methodologies, Phys Med Biol 2011.

3. Jenkinson, M., and S. Smith. "Medical Image Registration." Physics in Medicine & Biology 46.3 (2001): R1. https://doi.org/10.1088/0031-9155/46/3/001.

4. Wang, J., and M. Zhang. "DeepFLASH: An Efficient Network for Learning-Based Medical Image Registration." 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2020, https://doi.org/10.1109/CVPR42600.2020.00399.

5. Zhao, S., Y. Dong, E. Chang, et al. "Recursive Cascaded Networks for Unsupervised Medical Image Registration." 2019 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE 2020, https://doi.org/10.1109/ICCV.2019.00510.

6. Zhao, S., T. Lau, J. Luo, et al. "Unsupervised 3D End-to-End Medical Image Registration With Volume Tweening Network." IEEE Journal of Biomedical and Health Informatics 24.5 (2020): pp. 1394-1404. https://doi.org/10.1109/JBHI.2020.2985142.

7. Nakajima, Y., N. Kadoya, T. Kanai, et al. "Evaluation of the Effect of User-Guided Deformable Image Registration of Thoracic Images on Registration Accuracy Among Users." Medical Dosimetry 45.3 (2020), https://doi.org/10.1016/j.meddos.2020.03.003.

8. Liu, Q., X. Xiang, J. Qin, et al. "Coverless Steganography Based on Image Retrieval of DenseNet Features and DWT Sequence Mapping." Knowledge-Based Systems 192 (2020): 105375. https://doi.org/10.1016/j.knosys.2020.105375

9. Boveiri, H. R., R. Khayami, R. Javidan, et al. "Medical Image Registration Using Deep Neural Networks: A Comprehensive Review." Computers & Electrical Engineering 87 (2020): 106767. https://doi.org/10.1016/j.compeleceng.2020.106767.

10. Zhao, J., Y. Zhou, J. Zhao, et al. "Rapid-Precision Position Measurement of Linear Motor Mover Based on Joint Spatial Phase Method." IEEE Transactions on Industrial Informatics 16.7 (2020): pp. 4333-4343. https://doi.org/10.1109/TII.2019.2956067.

11. Singh, S., and R. S. Anand. "Multimodal Medical Image Fusion Using Hybrid Layer Decomposition With CNN-Based Feature Mapping and Structural Clustering." IEEE Transactions on Instrumentation and Measurement 69.6 (2020): pp. 3855-3865. https://doi.org/10.1109/TIM.2020.2969923.

12. Roy, S., and P. Maji. "Medical Image Segmentation by Partitioning Spatially Constrained Fuzzy Approximation Spaces." IEEE Transactions on Fuzzy Systems 28.5 (2020): pp. 965-977. https://doi.org/10.1109/TFUZZ.2019.2946769.

13. Dolly, D. R. J., J. D. Peter, G. J. Bala, et al. "Image Fusion for Stabilized Medical Video Sequence Using Multimodal Parametric Registration." Pattern Recognition Letters (2020), https://doi.org/10.1016/j.patrec.2020.05.010.

14. A. A. P., B. A. S., and C. K. B. "Compression and Multiplexing of Medical Images Using Optical Image Processing." Computational Intelligence and Its Applications in Healthcare (2020): pp. 63-71.