Extended nnU-Net for Brain Metastasis Detection and Segmentation in Contrast-Enhanced Magnetic Resonance Imaging With a Large Multi-Institutional Data Set.

IF 6.4 1区 医学 Q1 ONCOLOGY
Youngjin Yoo, Eli Gibson, Gengyan Zhao, Thomas J Re, Hemant Parmar, Jyotipriya Das, Hesheng Wang, Michelle M Kim, Colette Shen, Yueh Lee, Douglas Kondziolka, Mohannad Ibrahim, Jun Lian, Rajan Jain, Tong Zhu, Dorin Comaniciu, James M Balter, Yue Cao
{"title":"Extended nnU-Net for Brain Metastasis Detection and Segmentation in Contrast-Enhanced Magnetic Resonance Imaging With a Large Multi-Institutional Data Set.","authors":"Youngjin Yoo, Eli Gibson, Gengyan Zhao, Thomas J Re, Hemant Parmar, Jyotipriya Das, Hesheng Wang, Michelle M Kim, Colette Shen, Yueh Lee, Douglas Kondziolka, Mohannad Ibrahim, Jun Lian, Rajan Jain, Tong Zhu, Dorin Comaniciu, James M Balter, Yue Cao","doi":"10.1016/j.ijrobp.2024.07.2318","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to investigate an extended self-adapting nnU-Net framework for detecting and segmenting brain metastases (BM) on magnetic resonance imaging (MRI).</p><p><strong>Methods and materials: </strong>Six different nnU-Net systems with adaptive data sampling, adaptive Dice loss, or different patch/batch sizes were trained and tested for detecting and segmenting intraparenchymal BM with a size ≥2 mm on 3 Dimensional (3D) post-Gd T1-weighted MRI volumes using 2092 patients from 7 institutions (1712, 195, and 185 patients for training, validation, and testing, respectively). Gross tumor volumes of BM delineated by physicians for stereotactic radiosurgery were collected retrospectively and curated at each institute. Additional centralized data curation was carried out to create gross tumor volumes of uncontoured BM by 2 radiologists to improve the accuracy of ground truth. The training data set was augmented with synthetic BMs of 1025 MRI volumes using a 3D generative pipeline. BM detection was evaluated by lesion-level sensitivity and false-positive (FP) rate. BM segmentation was assessed by lesion-level Dice similarity coefficient, 95-percentile Hausdorff distance, and average Hausdorff distance (HD). The performances were assessed across different BM sizes. Additional testing was performed using a second data set of 206 patients.</p><p><strong>Results: </strong>Of the 6 nnU-Net systems, the nnU-Net with adaptive Dice loss achieved the best detection and segmentation performance on the first testing data set. At an FP rate of 0.65 ± 1.17, overall sensitivity was 0.904 for all sizes of BM, 0.966 for BM ≥0.1 cm<sup>3</sup>, and 0.824 for BM <0.1 cm<sup>3</sup>. Mean values of Dice similarity coefficient, 95-percentile Hausdorff distance, and average HD of all detected BMs were 0.758, 1.45, and 0.23 mm, respectively. Performances on the second testing data set achieved a sensitivity of 0.907 at an FP rate of 0.57 ± 0.85 for all BM sizes, and an average HD of 0.33 mm for all detected BM.</p><p><strong>Conclusions: </strong>Our proposed extension of the self-configuring nnU-Net framework substantially improved small BM detection sensitivity while maintaining a controlled FP rate. Clinical utility of the extended nnU-Net model for assisting early BM detection and stereotactic radiosurgery planning will be investigated.</p>","PeriodicalId":14215,"journal":{"name":"International Journal of Radiation Oncology Biology Physics","volume":" ","pages":"241-249"},"PeriodicalIF":6.4000,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Radiation Oncology Biology Physics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1016/j.ijrobp.2024.07.2318","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/7/25 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"ONCOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Purpose: The purpose of this study was to investigate an extended self-adapting nnU-Net framework for detecting and segmenting brain metastases (BM) on magnetic resonance imaging (MRI).

Methods and materials: Six different nnU-Net systems with adaptive data sampling, adaptive Dice loss, or different patch/batch sizes were trained and tested for detecting and segmenting intraparenchymal BM with a size ≥2 mm on 3 Dimensional (3D) post-Gd T1-weighted MRI volumes using 2092 patients from 7 institutions (1712, 195, and 185 patients for training, validation, and testing, respectively). Gross tumor volumes of BM delineated by physicians for stereotactic radiosurgery were collected retrospectively and curated at each institute. Additional centralized data curation was carried out to create gross tumor volumes of uncontoured BM by 2 radiologists to improve the accuracy of ground truth. The training data set was augmented with synthetic BMs of 1025 MRI volumes using a 3D generative pipeline. BM detection was evaluated by lesion-level sensitivity and false-positive (FP) rate. BM segmentation was assessed by lesion-level Dice similarity coefficient, 95-percentile Hausdorff distance, and average Hausdorff distance (HD). The performances were assessed across different BM sizes. Additional testing was performed using a second data set of 206 patients.

Results: Of the 6 nnU-Net systems, the nnU-Net with adaptive Dice loss achieved the best detection and segmentation performance on the first testing data set. At an FP rate of 0.65 ± 1.17, overall sensitivity was 0.904 for all sizes of BM, 0.966 for BM ≥0.1 cm3, and 0.824 for BM <0.1 cm3. Mean values of Dice similarity coefficient, 95-percentile Hausdorff distance, and average HD of all detected BMs were 0.758, 1.45, and 0.23 mm, respectively. Performances on the second testing data set achieved a sensitivity of 0.907 at an FP rate of 0.57 ± 0.85 for all BM sizes, and an average HD of 0.33 mm for all detected BM.

Conclusions: Our proposed extension of the self-configuring nnU-Net framework substantially improved small BM detection sensitivity while maintaining a controlled FP rate. Clinical utility of the extended nnU-Net model for assisting early BM detection and stereotactic radiosurgery planning will be investigated.

利用大型多机构数据集在对比增强磁共振成像中进行脑转移瘤检测和分割的扩展 nnU-Net。
目的:研究一种扩展的自适应 nnU-Net 框架,用于在核磁共振成像上检测和分割脑转移瘤(BM):采用自适应数据采样、自适应骰子损失(ADL)或不同补丁/批量大小的六种不同 nnU-Net 系统进行训练和测试,利用来自七家机构的 2092 名患者(分别有 1712、195 和 185 名患者接受训练、验证和测试)在三维钆后 T1 加权磁共振成像卷上检测和分割大小≥ 2 毫米的实质内肿瘤。医生在进行立体定向放射外科手术(SRS)时划定的 BM 总肿瘤体积(GTV)由各机构进行回顾性收集和整理。另外,两名放射科医生还对数据进行了集中管理,以创建未经检查的 BM 的 GTV,从而提高地面实况的准确性。使用三维生成管道,用 1025 个 MRI 卷的合成 BM 增强了训练数据集。通过病灶级灵敏度和假阳性(FP)率来评估 BM 检测。通过病灶级 Dice 相似系数 (DSC)、95 百分位数 Hausdorff 距离 (HD95) 和平均 HD 来评估 BM 分割。对不同大小的 BM 的性能进行了评估。此外,还使用包含 206 名患者的第二个数据集进行了测试:结果:在六个 nnU-Net 系统中,带有 ADL 的 nnU-Net 在第一个测试数据集上的检测和分割性能最好。在 FP 率为 0.65±1.17 时,所有大小的 BM 的总体灵敏度均为 0.904,BM ≥ 0.1 cm3 为 0.966,BM < 0.1 cm3 为 0.824。所有检测到的 BM 的 DSC、HD95 和平均 HD 的平均值分别为 0.758、1.45 毫米和 0.23 毫米。第二个测试数据集的灵敏度为 0.907,所有 BM 大小的 FP 率为 0.57±0.85,所有检测到的 BM 的平均 HD 为 0.33 毫米:我们提出的自配置 nnU-Net 框架扩展大大提高了小型 BM 的检测灵敏度,同时保持了可控的 FP 率。我们将研究扩展 nnU-Net 模型在辅助早期 BM 检测和 SRS 计划方面的临床实用性。
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
CiteScore
11.00
自引率
7.10%
发文量
2538
审稿时长
6.6 weeks
期刊介绍: International Journal of Radiation Oncology • Biology • Physics (IJROBP), known in the field as the Red Journal, publishes original laboratory and clinical investigations related to radiation oncology, radiation biology, medical physics, and both education and health policy as it relates to the field. This journal has a particular interest in original contributions of the following types: prospective clinical trials, outcomes research, and large database interrogation. In addition, it seeks reports of high-impact innovations in single or combined modality treatment, tumor sensitization, normal tissue protection (including both precision avoidance and pharmacologic means), brachytherapy, particle irradiation, and cancer imaging. Technical advances related to dosimetry and conformal radiation treatment planning are of interest, as are basic science studies investigating tumor physiology and the molecular biology underlying cancer and normal tissue radiation response.
×
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