Fully automated grading of pituitary adenoma

Q4 Neuroscience
Raffaele Da Mutten , Olivier Zanier , Massimo Bottini , Yves Baumann , Olga Ciobanu-Caraus , Luca Regli , Carlo Serra , Victor E. Staartjes
{"title":"Fully automated grading of pituitary adenoma","authors":"Raffaele Da Mutten ,&nbsp;Olivier Zanier ,&nbsp;Massimo Bottini ,&nbsp;Yves Baumann ,&nbsp;Olga Ciobanu-Caraus ,&nbsp;Luca Regli ,&nbsp;Carlo Serra ,&nbsp;Victor E. Staartjes","doi":"10.1016/j.ynirp.2025.100233","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>The Zurich Pituitary Score (ZPS) is an externally validated radiological grading scale to predict the likelihood of gross total resection (GTR) on coronal T1w magnetic resonance imaging of pituitary adenomas. The ZPS is based on the ratio of maximum tumor horizontal diameter and minimum intercarotid distance and on carotid artery encasement. While the interobserver agreement of the ZPS was relatively good, automated grading would be beneficial.</div></div><div><h3>Methods</h3><div>A nnU-Net algorithm was trained to segment the manually labeled tumor tissue and the cavernous segment of the internal carotid artery. Subsequently, maximum horizontal tumor diameter and minimum intercarotid distance were extracted. Last, a seed-growing algorithm checked for encasement of the carotid to determine the ZPS.</div></div><div><h3>Results</h3><div>213 patients were included, of which 128 (60%) had non-functioning adenomas, 49 (23%) a growth-hormone secreting and 19 (9%) a prolactin producing tumor. Accordingly, ZPS gradings were I = 63 (30%), II = 94 (44%), III = 41 (19%) and IV = 15 (7%). Dice score (mean ± standard deviation) for the tumor, left carotid, and right carotid in training validation of 0.78 ± 0.24, 0.62 ± 0.31, 0.62 ± 0.30 and during holdout testing of 0.79 ± 0.24, 0.59 ± 0.32, 0.58 ± 0.33 was reached. After the exclusion of two cases with poor segmentation results, intraclass correlation coefficients [95% CI] for the intercarotid distance, maximum horizontal tumor diameter, and the ZPS ratio of the two measurements were 0.89 [0.80, 0.94], 0.91 [0.82, 0.96], 0.80 [0.66, 0.89] respectively. Cohen's weighted Kappa for the final ZPS grading was 0.79 [0.68, 0.90] and Spearman rank correlation was 0.83.</div></div><div><h3>Conclusions</h3><div>We developed and internally validated a machine learning-based method for fully automated grading of the ZPS. Generally, robust segmentation performance was achieved. While ZPS grading generally worked well, human ratings remain superior in many situations. Especially for raters with low experience, our approach offers a solid and objective alternative.</div></div>","PeriodicalId":74277,"journal":{"name":"Neuroimage. Reports","volume":"5 1","pages":"Article 100233"},"PeriodicalIF":0.0000,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neuroimage. Reports","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666956025000017","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Neuroscience","Score":null,"Total":0}
引用次数: 0

Abstract

Background

The Zurich Pituitary Score (ZPS) is an externally validated radiological grading scale to predict the likelihood of gross total resection (GTR) on coronal T1w magnetic resonance imaging of pituitary adenomas. The ZPS is based on the ratio of maximum tumor horizontal diameter and minimum intercarotid distance and on carotid artery encasement. While the interobserver agreement of the ZPS was relatively good, automated grading would be beneficial.

Methods

A nnU-Net algorithm was trained to segment the manually labeled tumor tissue and the cavernous segment of the internal carotid artery. Subsequently, maximum horizontal tumor diameter and minimum intercarotid distance were extracted. Last, a seed-growing algorithm checked for encasement of the carotid to determine the ZPS.

Results

213 patients were included, of which 128 (60%) had non-functioning adenomas, 49 (23%) a growth-hormone secreting and 19 (9%) a prolactin producing tumor. Accordingly, ZPS gradings were I = 63 (30%), II = 94 (44%), III = 41 (19%) and IV = 15 (7%). Dice score (mean ± standard deviation) for the tumor, left carotid, and right carotid in training validation of 0.78 ± 0.24, 0.62 ± 0.31, 0.62 ± 0.30 and during holdout testing of 0.79 ± 0.24, 0.59 ± 0.32, 0.58 ± 0.33 was reached. After the exclusion of two cases with poor segmentation results, intraclass correlation coefficients [95% CI] for the intercarotid distance, maximum horizontal tumor diameter, and the ZPS ratio of the two measurements were 0.89 [0.80, 0.94], 0.91 [0.82, 0.96], 0.80 [0.66, 0.89] respectively. Cohen's weighted Kappa for the final ZPS grading was 0.79 [0.68, 0.90] and Spearman rank correlation was 0.83.

Conclusions

We developed and internally validated a machine learning-based method for fully automated grading of the ZPS. Generally, robust segmentation performance was achieved. While ZPS grading generally worked well, human ratings remain superior in many situations. Especially for raters with low experience, our approach offers a solid and objective alternative.
垂体腺瘤全自动分级
背景:苏黎世垂体评分(ZPS)是一种经外部验证的放射学分级量表,用于预测垂体腺瘤冠状T1w磁共振成像中大体全切除(GTR)的可能性。ZPS是基于最大肿瘤水平直径与最小颈动脉间距的比值和颈动脉包裹情况。虽然ZPS的观察员间协议相对较好,但自动分级将是有益的。方法采用nnU-Net算法对人工标记的肿瘤组织和颈内动脉海绵样段进行分割。随后,提取最大水平肿瘤直径和最小颈动脉间距。最后,一个种子生长算法检查颈动脉的包膜,以确定ZPS。结果共纳入213例患者,其中128例(60%)为无功能腺瘤,49例(23%)为生长激素分泌瘤,19例(9%)为泌乳素分泌瘤。因此,保证各类I = 63 (30%), 2 = 94 (44%), 3 = 41 (19%), IV = 15(7%)。训练验证时肿瘤、左颈动脉和右颈动脉的Dice评分(平均值±标准差)分别为0.78±0.24、0.62±0.31、0.62±0.30,对照组测试时分别为0.79±0.24、0.59±0.32、0.58±0.33。在排除两例分割效果较差的病例后,颈动脉间距、最大水平肿瘤直径和两项测量的类内相关系数[95% CI]分别为0.89[0.80,0.94]、0.91[0.82,0.96]、0.80[0.66,0.89]。最终ZPS评分的Cohen加权Kappa为0.79 [0.68,0.90],Spearman秩相关系数为0.83。我们开发并内部验证了一种基于机器学习的ZPS全自动分级方法。总体而言,实现了鲁棒的分割性能。虽然ZPS评分通常效果很好,但在许多情况下,人类评分仍然更好。特别是对于经验较低的评级员,我们的方法提供了一个可靠和客观的选择。
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
Neuroimage. Reports
Neuroimage. Reports Neuroscience (General)
CiteScore
1.90
自引率
0.00%
发文量
0
审稿时长
87 days
×
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:604180095
Book学术官方微信