Brain Magnetic Resonance Image Inpainting via Deep Edge Region-based Generative Adversarial Network

IF 1.6 4区 工程技术 Q3 ENGINEERING, ELECTRICAL & ELECTRONIC
R. Kala, Raja Chandrasekaran, A. Ahilan, P. Jayapriya
{"title":"Brain Magnetic Resonance Image Inpainting via Deep Edge Region-based Generative Adversarial Network","authors":"R. Kala, Raja Chandrasekaran, A. Ahilan, P. Jayapriya","doi":"10.1007/s42835-024-02025-0","DOIUrl":null,"url":null,"abstract":"<p>Human brains are the most complex organs. There are a number of functions that this three-pound organ performs, including intelligence, interpreter of the senses, initiator of bodily movements, and controller of behaviour. In this paper, a novel ER-GAN model has been proposed for image inpainting (IIP) Brain MRI images. Initially, the brain MRI images are segmented using Attention V-Net. In the first GAN, Edge reconstruction Generative Adversarial Networks (EGAN) are used as edge generators able to hallucinate edges in missing regions based on the rest of the image’s edges and grayscale pixel intensities. Edge generation in brain MRI images involves leveraging these grayscale pixel intensities to detect boundaries between different brain tissues or structures. The varying intensities in MRI images often correspond to changes in tissue composition or boundaries between anatomical regions, making them valuable for edge detection and delineation. The second GAN uses the Region Reconstruction Generative Adversarial Network (RGAN) to fill in the missing regions by combining edge information from the missing regions and color and texture information from the surrounding regions. In experimental analysis, the Jaccard Index (JI) and Dice Index (DI) are obtained at 0.78 and 0.84 respectively. The proposed ER-GAN model reaches an overall accuracy of 99.25%, which is comparatively better than the existing techniques.</p>","PeriodicalId":15577,"journal":{"name":"Journal of Electrical Engineering & Technology","volume":null,"pages":null},"PeriodicalIF":1.6000,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Electrical Engineering & Technology","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s42835-024-02025-0","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

Human brains are the most complex organs. There are a number of functions that this three-pound organ performs, including intelligence, interpreter of the senses, initiator of bodily movements, and controller of behaviour. In this paper, a novel ER-GAN model has been proposed for image inpainting (IIP) Brain MRI images. Initially, the brain MRI images are segmented using Attention V-Net. In the first GAN, Edge reconstruction Generative Adversarial Networks (EGAN) are used as edge generators able to hallucinate edges in missing regions based on the rest of the image’s edges and grayscale pixel intensities. Edge generation in brain MRI images involves leveraging these grayscale pixel intensities to detect boundaries between different brain tissues or structures. The varying intensities in MRI images often correspond to changes in tissue composition or boundaries between anatomical regions, making them valuable for edge detection and delineation. The second GAN uses the Region Reconstruction Generative Adversarial Network (RGAN) to fill in the missing regions by combining edge information from the missing regions and color and texture information from the surrounding regions. In experimental analysis, the Jaccard Index (JI) and Dice Index (DI) are obtained at 0.78 and 0.84 respectively. The proposed ER-GAN model reaches an overall accuracy of 99.25%, which is comparatively better than the existing techniques.

Abstract Image

通过基于边缘区域的深度生成对抗网络绘制脑磁共振图像
人类的大脑是最复杂的器官。这个三磅重的器官具有多种功能,包括智能、感官解释器、身体运动的启动器和行为控制器。本文提出了一种新颖的 ER-GAN 模型,用于对脑核磁共振成像图像进行图像着色 (IIP)。首先,使用注意力 V-Net 对大脑 MRI 图像进行分割。在第一个 GAN 中,边缘重构生成对抗网络(EGAN)被用作边缘生成器,能够根据图像的其他边缘和灰度像素强度在缺失区域生成边缘。脑部核磁共振成像图像中的边缘生成包括利用这些灰度像素强度来检测不同脑组织或结构之间的边界。核磁共振成像图像中的不同强度往往对应着组织成分的变化或解剖区域之间的边界,因此对边缘检测和划分非常有价值。第二个 GAN 使用区域重建生成对抗网络(RGAN),结合缺失区域的边缘信息和周围区域的颜色和纹理信息来填补缺失区域。在实验分析中,Jaccard 指数(JI)和 Dice 指数(DI)分别为 0.78 和 0.84。所提出的 ER-GAN 模型的总体准确率达到 99.25%,相对优于现有技术。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Journal of Electrical Engineering & Technology
Journal of Electrical Engineering & Technology ENGINEERING, ELECTRICAL & ELECTRONIC-
CiteScore
4.00
自引率
15.80%
发文量
321
审稿时长
3.8 months
期刊介绍: ournal of Electrical Engineering and Technology (JEET), which is the official publication of the Korean Institute of Electrical Engineers (KIEE) being published bimonthly, released the first issue in March 2006.The journal is open to submission from scholars and experts in the wide areas of electrical engineering technologies. The scope of the journal includes all issues in the field of Electrical Engineering and Technology. Included are techniques for electrical power engineering, electrical machinery and energy conversion systems, electrophysics and applications, information and controls.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信