利用贝叶斯惩罚似然算法对重建图像进行深度训练以增强 PET 图像的效果

IF 1.6 4区 医学 Q4 ENGINEERING, BIOMEDICAL
Ali Ghafari, Mahsa Shahrbabaki Mofrad, Nima Kasraie, Mohammad Reza Ay, Negisa Seyyedi, Peyman Sheikhzadeh
{"title":"利用贝叶斯惩罚似然算法对重建图像进行深度训练以增强 PET 图像的效果","authors":"Ali Ghafari, Mahsa Shahrbabaki Mofrad, Nima Kasraie, Mohammad Reza Ay, Negisa Seyyedi, Peyman Sheikhzadeh","doi":"10.1007/s40846-024-00882-8","DOIUrl":null,"url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Purpose</h3><p>To adopt the merits of the Bayesian Penalized Likelihood (BPL) reconstruction algorithm (incl. improved contrast recovery), a deep learning ResNet model was trained to estimate BPL-like images using the non-attenuation, non-scatter corrected PET images (PET-nonAC) as inputs.</p><h3 data-test=\"abstract-sub-heading\">Methods</h3><p>Images of 112 patients were used for model training (79 patients), validation (13 patients) and testing (20 patients). The ResNet model used PET-nonAC images as input and predicted corresponding BPL-like images. The model performance regarding image quality was evaluated using metrics such as contrast-to-noise ratio (CNR).</p><h3 data-test=\"abstract-sub-heading\">Results</h3><p>The CNR of the reference BPL images was 2.40, while estimated BPL-like images using the deep learning model have a CNR value of 2.42 indicative of comparable performance.</p><h3 data-test=\"abstract-sub-heading\">Conclusion</h3><p>The estimated BPL-like images of the deep learning model offer comparable quality to the reference BPL images especially regarding the CNR metric. This deep learning model can be used to improve the image quality PET-nonAC by adopting the characteristics of the BPL images.</p>","PeriodicalId":50133,"journal":{"name":"Journal of Medical and Biological Engineering","volume":"24 1","pages":""},"PeriodicalIF":1.6000,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"PET Images Enhancement Using Deep Training of Reconstructed Images with Bayesian Penalized Likelihood Algorithm\",\"authors\":\"Ali Ghafari, Mahsa Shahrbabaki Mofrad, Nima Kasraie, Mohammad Reza Ay, Negisa Seyyedi, Peyman Sheikhzadeh\",\"doi\":\"10.1007/s40846-024-00882-8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<h3 data-test=\\\"abstract-sub-heading\\\">Purpose</h3><p>To adopt the merits of the Bayesian Penalized Likelihood (BPL) reconstruction algorithm (incl. improved contrast recovery), a deep learning ResNet model was trained to estimate BPL-like images using the non-attenuation, non-scatter corrected PET images (PET-nonAC) as inputs.</p><h3 data-test=\\\"abstract-sub-heading\\\">Methods</h3><p>Images of 112 patients were used for model training (79 patients), validation (13 patients) and testing (20 patients). The ResNet model used PET-nonAC images as input and predicted corresponding BPL-like images. The model performance regarding image quality was evaluated using metrics such as contrast-to-noise ratio (CNR).</p><h3 data-test=\\\"abstract-sub-heading\\\">Results</h3><p>The CNR of the reference BPL images was 2.40, while estimated BPL-like images using the deep learning model have a CNR value of 2.42 indicative of comparable performance.</p><h3 data-test=\\\"abstract-sub-heading\\\">Conclusion</h3><p>The estimated BPL-like images of the deep learning model offer comparable quality to the reference BPL images especially regarding the CNR metric. This deep learning model can be used to improve the image quality PET-nonAC by adopting the characteristics of the BPL images.</p>\",\"PeriodicalId\":50133,\"journal\":{\"name\":\"Journal of Medical and Biological Engineering\",\"volume\":\"24 1\",\"pages\":\"\"},\"PeriodicalIF\":1.6000,\"publicationDate\":\"2024-07-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Medical and Biological Engineering\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1007/s40846-024-00882-8\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"ENGINEERING, BIOMEDICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Medical and Biological Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s40846-024-00882-8","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0

摘要

目的采用贝叶斯惩罚似然(BPL)重建算法的优点(包括改进的对比度恢复),训练一个深度学习ResNet模型,以非衰减、非散射校正的PET图像(PET-nonAC)为输入,估计BPL-like图像。方法使用112名患者的图像进行模型训练(79名患者)、验证(13名患者)和测试(20名患者)。ResNet 模型使用 PET-nonAC 图像作为输入,并预测相应的 BPL-like 图像。结果参考 BPL 图像的 CNR 为 2.40,而使用深度学习模型估计的 BPL 类图像的 CNR 值为 2.42,表明性能相当。这种深度学习模型可以通过采用 BPL 图像的特征来改善 PET-nonAC 的图像质量。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

PET Images Enhancement Using Deep Training of Reconstructed Images with Bayesian Penalized Likelihood Algorithm

PET Images Enhancement Using Deep Training of Reconstructed Images with Bayesian Penalized Likelihood Algorithm

Purpose

To adopt the merits of the Bayesian Penalized Likelihood (BPL) reconstruction algorithm (incl. improved contrast recovery), a deep learning ResNet model was trained to estimate BPL-like images using the non-attenuation, non-scatter corrected PET images (PET-nonAC) as inputs.

Methods

Images of 112 patients were used for model training (79 patients), validation (13 patients) and testing (20 patients). The ResNet model used PET-nonAC images as input and predicted corresponding BPL-like images. The model performance regarding image quality was evaluated using metrics such as contrast-to-noise ratio (CNR).

Results

The CNR of the reference BPL images was 2.40, while estimated BPL-like images using the deep learning model have a CNR value of 2.42 indicative of comparable performance.

Conclusion

The estimated BPL-like images of the deep learning model offer comparable quality to the reference BPL images especially regarding the CNR metric. This deep learning model can be used to improve the image quality PET-nonAC by adopting the characteristics of the BPL images.

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
CiteScore
4.30
自引率
5.00%
发文量
81
审稿时长
3 months
期刊介绍: The purpose of Journal of Medical and Biological Engineering, JMBE, is committed to encouraging and providing the standard of biomedical engineering. The journal is devoted to publishing papers related to clinical engineering, biomedical signals, medical imaging, bio-informatics, tissue engineering, and so on. Other than the above articles, any contributions regarding hot issues and technological developments that help reach the purpose are also included.
×
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