Zobrazit v EDS

A Quantitative Microwave Imaging Approach for Brain Stroke Classification Based on the Generalized Tikhonov Regularization

Uloženo v:
Podrobná bibliografie
Název: A Quantitative Microwave Imaging Approach for Brain Stroke Classification Based on the Generalized Tikhonov Regularization
Autoři: Sayyed Saleh Sayyed Mousavi, Mohammad Saeed Majedi
Zdroj: IEEE Access, Vol 11, Pp 73370-73376 (2023)
Informace o vydavateli: Institute of Electrical and Electronics Engineers (IEEE), 2023.
Rok vydání: 2023
Témata: Microwave imaging, Electrical engineering. Electronics. Nuclear engineering, generalized Tikhonov regularization, brain stroke classification, TK1-9971, 3. Good health
Popis: Early diagnosis of stroke type by imaging is one of the most important tasks for stroke patients. In this article, we propose a new approach to reconstruct the brain image with high accuracy and quality. In this approach, first we use the Born iterative method to reconstruct the brain image. Then by comparing this image with a set of MRI-based brain images, using structural similarity index measure criterion, we choose the best one as reference image. Finally, we reconstruct the brain image by distorted Born iterative method or Born iterative method along with generalized Tikhonov regularization using the reference image. The reconstructed images are compared with those that obtained based on Tikhonov regularization. These comparisons demonstrate that the accuracy and quality of images in the proposed approach are significantly increased.
Druh dokumentu: Article
ISSN: 2169-3536
DOI: 10.1109/access.2023.3295692
Přístupová URL adresa: https://doaj.org/article/7f95ef2700d948cc9a272738ab692dce
Rights: CC BY NC ND
Přístupové číslo: edsair.doi.dedup.....f4166c2d21cb450c8360691dfeb84c9e
Databáze: OpenAIRE
Popis
Abstrakt:Early diagnosis of stroke type by imaging is one of the most important tasks for stroke patients. In this article, we propose a new approach to reconstruct the brain image with high accuracy and quality. In this approach, first we use the Born iterative method to reconstruct the brain image. Then by comparing this image with a set of MRI-based brain images, using structural similarity index measure criterion, we choose the best one as reference image. Finally, we reconstruct the brain image by distorted Born iterative method or Born iterative method along with generalized Tikhonov regularization using the reference image. The reconstructed images are compared with those that obtained based on Tikhonov regularization. These comparisons demonstrate that the accuracy and quality of images in the proposed approach are significantly increased.
ISSN:21693536
DOI:10.1109/access.2023.3295692