Deep learning-based segmentation of kidneys and renal cysts on T2-weighted MRI from patients with autosomal dominant polycystic kidney disease.

IF 3.7 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Rémi Sore, Pascal Cathier, Anna Sesilia Vlachomitrou, Jérôme Bailleux, Karine Arnaud, Laurent Juillard, Sandrine Lemoine, Olivier Rouvière
{"title":"Deep learning-based segmentation of kidneys and renal cysts on T2-weighted MRI from patients with autosomal dominant polycystic kidney disease.","authors":"Rémi Sore, Pascal Cathier, Anna Sesilia Vlachomitrou, Jérôme Bailleux, Karine Arnaud, Laurent Juillard, Sandrine Lemoine, Olivier Rouvière","doi":"10.1186/s41747-024-00520-7","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Our aim was to train and test a deep learning-based algorithm for automatically segmenting kidneys and renal cysts in patients with autosomal dominant polycystic kidney disease (ADPKD).</p><p><strong>Methods: </strong>We retrospectively selected all ADPKD patients who underwent renal MRI with coronal T2-weighted imaging at our institution from 2008 to 2022. The 20 most recent examinations constituted the test dataset, to mimic pseudoprospective enrolment. The remaining ones constituted the training dataset to which eight normal renal MRIs were added. Kidneys and cysts ground truth segmentations were performed on coronal T2-weighted images by a junior radiologist supervised by an experienced radiologist. Kidneys and cysts of the 20 test MRIs were segmented by the algorithm and three independent human raters. Segmentations were compared using overlap metrics. The total kidney volume (TKV), total cystic volume (TCV), and cystic index (TCV divided by TKV) were compared using Bland-Altman analysis.</p><p><strong>Results: </strong>We included 164 ADPKD patients. Dice similarity coefficients ranged from 85.9% to 87.4% between the algorithms and the raters' segmentations and from 84.2% to 86.2% across raters' segmentations. For TCV assessment, the biases ± standard deviations (SD) were 3-19 ± 137-151 mL between the algorithm and the raters, and 22-45 ± 49-57 mL across raters. The algorithm underestimated TKV and TCV in two outliers with TCV > 2800 mL. For cystic index assessment, the biases ± SD were 2.5-6.9% ± 6.7-8.3% between the algorithm and the raters, and 2.1-9.4 ± 7.4-11.6% across raters.</p><p><strong>Conclusion: </strong>The algorithm's performance fell within the range of inter-rater variability, but large TKV and TCV were underestimated.</p><p><strong>Relevance statement: </strong>Accurate automated segmentation of the renal cysts will enable the large-scale evaluation of the prognostic value of TCV and cystic index in ADPKD patients. If these biomarkers are prognostic, then automated segmentation will facilitate their use in daily routine.</p><p><strong>Key points: </strong>Cystic volume is an emerging biomarker in ADPKD. The algorithm's performance in segmenting kidneys and cysts fell within interrater variability. The segmentation of very large cysts, under-represented in the training dataset, needs improvement.</p>","PeriodicalId":36926,"journal":{"name":"European Radiology Experimental","volume":null,"pages":null},"PeriodicalIF":3.7000,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11525362/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Radiology Experimental","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1186/s41747-024-00520-7","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Background: Our aim was to train and test a deep learning-based algorithm for automatically segmenting kidneys and renal cysts in patients with autosomal dominant polycystic kidney disease (ADPKD).

Methods: We retrospectively selected all ADPKD patients who underwent renal MRI with coronal T2-weighted imaging at our institution from 2008 to 2022. The 20 most recent examinations constituted the test dataset, to mimic pseudoprospective enrolment. The remaining ones constituted the training dataset to which eight normal renal MRIs were added. Kidneys and cysts ground truth segmentations were performed on coronal T2-weighted images by a junior radiologist supervised by an experienced radiologist. Kidneys and cysts of the 20 test MRIs were segmented by the algorithm and three independent human raters. Segmentations were compared using overlap metrics. The total kidney volume (TKV), total cystic volume (TCV), and cystic index (TCV divided by TKV) were compared using Bland-Altman analysis.

Results: We included 164 ADPKD patients. Dice similarity coefficients ranged from 85.9% to 87.4% between the algorithms and the raters' segmentations and from 84.2% to 86.2% across raters' segmentations. For TCV assessment, the biases ± standard deviations (SD) were 3-19 ± 137-151 mL between the algorithm and the raters, and 22-45 ± 49-57 mL across raters. The algorithm underestimated TKV and TCV in two outliers with TCV > 2800 mL. For cystic index assessment, the biases ± SD were 2.5-6.9% ± 6.7-8.3% between the algorithm and the raters, and 2.1-9.4 ± 7.4-11.6% across raters.

Conclusion: The algorithm's performance fell within the range of inter-rater variability, but large TKV and TCV were underestimated.

Relevance statement: Accurate automated segmentation of the renal cysts will enable the large-scale evaluation of the prognostic value of TCV and cystic index in ADPKD patients. If these biomarkers are prognostic, then automated segmentation will facilitate their use in daily routine.

Key points: Cystic volume is an emerging biomarker in ADPKD. The algorithm's performance in segmenting kidneys and cysts fell within interrater variability. The segmentation of very large cysts, under-represented in the training dataset, needs improvement.

基于深度学习的常染色体显性多囊肾患者 T2 加权核磁共振成像上的肾脏和肾囊肿分割。
背景:我们的目的是训练和测试一种基于深度学习的算法,用于自动分割常染色体显性多囊肾(ADPKD)患者的肾脏和肾囊肿:我们回顾性地选择了2008年至2022年期间在我院接受冠状T2加权成像肾脏核磁共振检查的所有ADPKD患者。最近的 20 次检查构成测试数据集,以模拟伪回顾性登记。其余的构成训练数据集,并在此基础上添加 8 个正常的肾脏 MRI。一名初级放射科医生在一名经验丰富的放射科医生的指导下,对冠状 T2 加权图像上的肾脏和囊肿进行地面实况分割。20 张测试核磁共振成像的肾脏和囊肿由算法和三位独立的人类评分员进行分割。使用重叠度量对分割结果进行比较。使用Bland-Altman分析比较肾脏总体积(TKV)、囊肿总体积(TCV)和囊肿指数(TCV除以TKV):我们共纳入了 164 名 ADPKD 患者。算法与评分者分段之间的骰子相似系数从85.9%到87.4%不等,评分者分段之间的相似系数从84.2%到86.2%不等。在 TCV 评估中,算法与评分者之间的偏差(± 标准差,SD)为 3-19 ± 137-151 mL,不同评分者之间的偏差(± 标准差,SD)为 22-45 ± 49-57 mL。在 TCV > 2800 mL 的两个异常值中,算法低估了 TKV 和 TCV。在囊肿指数评估方面,算法与评分者之间的偏差(± SD)为 2.5-6.9% ± 6.7-8.3%,不同评分者之间的偏差(± SD)为 2.1-9.4 ± 7.4-11.6%:结论:该算法的性能在评分者之间的变异范围内,但大TKV和TCV被低估了:对肾囊肿进行准确的自动分割将有助于大规模评估 TCV 和囊肿指数在 ADPKD 患者中的预后价值。如果这些生物标志物具有预后价值,那么自动分割将有助于它们在日常工作中的应用:囊肿体积是ADPKD的一种新兴生物标志物。该算法在分割肾脏和囊肿方面的表现在评定者之间存在差异。超大囊肿在训练数据集中所占比例较低,因此需要改进对超大囊肿的分割。
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
European Radiology Experimental
European Radiology Experimental Medicine-Radiology, Nuclear Medicine and Imaging
CiteScore
6.70
自引率
2.60%
发文量
56
审稿时长
18 weeks
×
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