基于深度学习的医学图像融合——基于概率参数导向图像滤波器的集成联合斜率分析

E. S. Rao, C. Prasad
{"title":"基于深度学习的医学图像融合——基于概率参数导向图像滤波器的集成联合斜率分析","authors":"E. S. Rao, C. Prasad","doi":"10.1142/s0219467822400137","DOIUrl":null,"url":null,"abstract":"Medical image fusion plays a significant role in medical diagnosis applications. Although the conventional approaches have produced moderate visual analysis, still there is a scope to improve the performance parameters and reduce the computational complexity. Thus, this article implemented the hybrid fusion method by using the novel implementation of joint slope analysis (JSA), probabilistic parametric steered image filtration (PPSIF), and deep learning convolutional neural networks (DLCNNs)-based SR Fusion Net. Here, JSA decomposes the images to estimate edge-based slopes and develops the edge-preserving approximate layers from the multi-modal medical images. Further, PPSIF is used to generate the feature fusion with base layer-based weight maps. Then, the SR Fusion Net is used to generate the spatial and texture feature-based weight maps. Finally, optimal fusion rule is applied on the detail layers generated from the base layer and approximate layer, which resulted in the fused outcome. The proposed method is capable of performing the fusion operation between various modalities of images, such as MRI-CT, MRI-PET, and MRI-SPECT combinations by using two different architectures. The simulation results show that the proposed method resulted in better subjective and objective performance as compared to state of art approaches.","PeriodicalId":177479,"journal":{"name":"Int. J. Image Graph.","volume":"89 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Deep Learning-Based Medical Image Fusion Using Integrated Joint Slope Analysis with Probabilistic Parametric Steered Image Filter\",\"authors\":\"E. S. Rao, C. Prasad\",\"doi\":\"10.1142/s0219467822400137\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Medical image fusion plays a significant role in medical diagnosis applications. Although the conventional approaches have produced moderate visual analysis, still there is a scope to improve the performance parameters and reduce the computational complexity. Thus, this article implemented the hybrid fusion method by using the novel implementation of joint slope analysis (JSA), probabilistic parametric steered image filtration (PPSIF), and deep learning convolutional neural networks (DLCNNs)-based SR Fusion Net. Here, JSA decomposes the images to estimate edge-based slopes and develops the edge-preserving approximate layers from the multi-modal medical images. Further, PPSIF is used to generate the feature fusion with base layer-based weight maps. Then, the SR Fusion Net is used to generate the spatial and texture feature-based weight maps. Finally, optimal fusion rule is applied on the detail layers generated from the base layer and approximate layer, which resulted in the fused outcome. The proposed method is capable of performing the fusion operation between various modalities of images, such as MRI-CT, MRI-PET, and MRI-SPECT combinations by using two different architectures. The simulation results show that the proposed method resulted in better subjective and objective performance as compared to state of art approaches.\",\"PeriodicalId\":177479,\"journal\":{\"name\":\"Int. J. Image Graph.\",\"volume\":\"89 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-04-08\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Int. J. Image Graph.\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1142/s0219467822400137\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Int. J. Image Graph.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1142/s0219467822400137","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

摘要

医学图像融合在医学诊断中有着重要的应用。虽然传统的方法已经产生了适度的可视化分析,但仍有改进性能参数和降低计算复杂度的余地。因此,本文采用联合斜率分析(JSA)、概率参数导向图像滤波(PPSIF)和基于深度学习卷积神经网络(dlcnn)的SR融合网络的新实现实现了混合融合方法。在这里,JSA对图像进行分解以估计基于边缘的斜率,并从多模态医学图像中开发保持边缘的近似层。在此基础上,利用PPSIF与基于基础层的权重图进行特征融合。然后,利用SR融合网生成基于空间和纹理特征的权重图。最后,对由基础层和近似层生成的细节层应用最优融合规则,得到融合结果。该方法通过使用两种不同的结构,能够在不同模式的图像之间进行融合操作,例如MRI-CT, MRI-PET和MRI-SPECT组合。仿真结果表明,与现有方法相比,该方法具有更好的主客观性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Deep Learning-Based Medical Image Fusion Using Integrated Joint Slope Analysis with Probabilistic Parametric Steered Image Filter
Medical image fusion plays a significant role in medical diagnosis applications. Although the conventional approaches have produced moderate visual analysis, still there is a scope to improve the performance parameters and reduce the computational complexity. Thus, this article implemented the hybrid fusion method by using the novel implementation of joint slope analysis (JSA), probabilistic parametric steered image filtration (PPSIF), and deep learning convolutional neural networks (DLCNNs)-based SR Fusion Net. Here, JSA decomposes the images to estimate edge-based slopes and develops the edge-preserving approximate layers from the multi-modal medical images. Further, PPSIF is used to generate the feature fusion with base layer-based weight maps. Then, the SR Fusion Net is used to generate the spatial and texture feature-based weight maps. Finally, optimal fusion rule is applied on the detail layers generated from the base layer and approximate layer, which resulted in the fused outcome. The proposed method is capable of performing the fusion operation between various modalities of images, such as MRI-CT, MRI-PET, and MRI-SPECT combinations by using two different architectures. The simulation results show that the proposed method resulted in better subjective and objective performance as compared to state of art approaches.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
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