What patient positioning in chest X-ray is still acceptable? – An empirical study on thresholding quality metrics

Imaging Pub Date : 2024-05-08 DOI:10.1556/1647.2024.00187
Jens von Berg, Kenneth F. M. Hergaarden, Max Englmaier, Daniela Pfeiffer, N. Wieberneit, Sven Krönke-Hille, T. Harder, André Gooßen, Daniel Bystrov, Matthias Brueck, Stewart Young, Hildo. J. Lamb
{"title":"What patient positioning in chest X-ray is still acceptable? – An empirical study on thresholding quality metrics","authors":"Jens von Berg, Kenneth F. M. Hergaarden, Max Englmaier, Daniela Pfeiffer, N. Wieberneit, Sven Krönke-Hille, T. Harder, André Gooßen, Daniel Bystrov, Matthias Brueck, Stewart Young, Hildo. J. Lamb","doi":"10.1556/1647.2024.00187","DOIUrl":null,"url":null,"abstract":"Issues in patient positioning during chest X-ray (CXR) acquisition impair diagnostic quality and potentially increase radiation dose. Automated quality assessment was proposed to address this. Our objective is to determine thresholds on some quality control metrics following international guidelines, that represent expert knowledge and can be applied in a comprehensible and explainable AI approach for such an automatic quality assessment.An AI-method estimating collimation distance to the ribcage, balancing between both clavicle heads, and number of ribs above the diaphragm as metrics for collimation, rotation, and inhalation quality was applied on 64,315 posteroanterior CXR images from a public dataset (ChestX-ray8). From this set 920 CXR images were sampled and manually annotated to gain additional trusted reference metrics. Seven readers from different institutions then classified the acquisition quality of these images independently into okay, inadequate, or unacceptable following the criteria of international guidelines. Optimal thresholds on the metrics were determined to reproduce these classes using the metrics only.A fair to moderate agreement between the experts was found. When disregarding all inadequate rates a classification on the metrics was able to separate okay rated cases from unacceptable cases for collimation (AUC 0.97), rotation (AUC = 0.93) and inhalation (AUC = 0.97).Suitable thresholds were determined to reproduce expert opinions in the assessment of the most important quality criteria in CXR acquisition. These thresholds were finally applied on the AI-method's estimates to automatically classify image acquisition quality comprehensibly and according to the guidelines.","PeriodicalId":503851,"journal":{"name":"Imaging","volume":" 4","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Imaging","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1556/1647.2024.00187","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Issues in patient positioning during chest X-ray (CXR) acquisition impair diagnostic quality and potentially increase radiation dose. Automated quality assessment was proposed to address this. Our objective is to determine thresholds on some quality control metrics following international guidelines, that represent expert knowledge and can be applied in a comprehensible and explainable AI approach for such an automatic quality assessment.An AI-method estimating collimation distance to the ribcage, balancing between both clavicle heads, and number of ribs above the diaphragm as metrics for collimation, rotation, and inhalation quality was applied on 64,315 posteroanterior CXR images from a public dataset (ChestX-ray8). From this set 920 CXR images were sampled and manually annotated to gain additional trusted reference metrics. Seven readers from different institutions then classified the acquisition quality of these images independently into okay, inadequate, or unacceptable following the criteria of international guidelines. Optimal thresholds on the metrics were determined to reproduce these classes using the metrics only.A fair to moderate agreement between the experts was found. When disregarding all inadequate rates a classification on the metrics was able to separate okay rated cases from unacceptable cases for collimation (AUC 0.97), rotation (AUC = 0.93) and inhalation (AUC = 0.97).Suitable thresholds were determined to reproduce expert opinions in the assessment of the most important quality criteria in CXR acquisition. These thresholds were finally applied on the AI-method's estimates to automatically classify image acquisition quality comprehensibly and according to the guidelines.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
在胸部 X 光检查中,什么样的患者定位仍可接受?- 关于阈值质量指标的实证研究
胸部 X 光(CXR)采集过程中患者定位的问题会影响诊断质量,并可能增加辐射剂量。为解决这一问题,我们提出了自动质量评估方法。我们的目标是根据国际指南确定一些质量控制指标的阈值,这些阈值代表了专家知识,可应用于自动质量评估的可理解和可解释的人工智能方法中。我们在公共数据集(ChestX-ray8)中的 64,315 张后正位 CXR 图像上应用了一种人工智能方法,该方法将到肋骨的准直距离、两个锁骨头之间的平衡以及横膈膜上方的肋骨数量作为准直、旋转和吸入质量的指标。从这组图像中抽取了 920 张 CXR 图像并进行了人工标注,以获得更多可信的参考指标。然后,来自不同机构的七位读者按照国际指南的标准,将这些图像的采集质量独立分为合格、不合格或不可接受。专家们的意见相当一致。在不考虑所有不合格率的情况下,指标分类能够将准直(AUC 0.97)、旋转(AUC = 0.93)和吸入(AUC = 0.97)方面的合格病例与不合格病例区分开来。最后将这些阈值应用于人工智能方法的估计值,以便根据指南对图像采集质量进行自动分类。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Testicular ultrasound measurement of testicular volume and epididymis diameters: Prediction of semen quality in a prospective study What patient positioning in chest X-ray is still acceptable? – An empirical study on thresholding quality metrics Pre-hospital ultrasonography screening for biliary disease in remote settings Evaluation of the cardiopulmonary size on slot-scanning full-spine radiographs Inter-reader reproducibility of a radiographic grading system for usual interstitial pneumonitis validates its use as a surrogate endpoint in clinical trials
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1