Hybrid representation-enhanced sampling for Bayesian active learning in musculoskeletal segmentation of lower extremities.

IF 2.3 3区 医学 Q3 ENGINEERING, BIOMEDICAL
Ganping Li, Yoshito Otake, Mazen Soufi, Masashi Taniguchi, Masahide Yagi, Noriaki Ichihashi, Keisuke Uemura, Masaki Takao, Nobuhiko Sugano, Yoshinobu Sato
{"title":"Hybrid representation-enhanced sampling for Bayesian active learning in musculoskeletal segmentation of lower extremities.","authors":"Ganping Li, Yoshito Otake, Mazen Soufi, Masashi Taniguchi, Masahide Yagi, Noriaki Ichihashi, Keisuke Uemura, Masaki Takao, Nobuhiko Sugano, Yoshinobu Sato","doi":"10.1007/s11548-024-03065-7","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Manual annotations for training deep learning models in auto-segmentation are time-intensive. This study introduces a hybrid representation-enhanced sampling strategy that integrates both density and diversity criteria within an uncertainty-based Bayesian active learning (BAL) framework to reduce annotation efforts by selecting the most informative training samples.</p><p><strong>Methods: </strong>The experiments are performed on two lower extremity datasets of MRI and CT images, focusing on the segmentation of the femur, pelvis, sacrum, quadriceps femoris, hamstrings, adductors, sartorius, and iliopsoas, utilizing a U-net-based BAL framework. Our method selects uncertain samples with high density and diversity for manual revision, optimizing for maximal similarity to unlabeled instances and minimal similarity to existing training data. We assess the accuracy and efficiency using dice and a proposed metric called reduced annotation cost (RAC), respectively. We further evaluate the impact of various acquisition rules on BAL performance and design an ablation study for effectiveness estimation.</p><p><strong>Results: </strong>In MRI and CT datasets, our method was superior or comparable to existing ones, achieving a 0.8% dice and 1.0% RAC increase in CT (statistically significant), and a 0.8% dice and 1.1% RAC increase in MRI (not statistically significant) in volume-wise acquisition. Our ablation study indicates that combining density and diversity criteria enhances the efficiency of BAL in musculoskeletal segmentation compared to using either criterion alone.</p><p><strong>Conclusion: </strong>Our sampling method is proven efficient in reducing annotation costs in image segmentation tasks. The combination of the proposed method and our BAL framework provides a semi-automatic way for efficient annotation of medical image datasets.</p>","PeriodicalId":51251,"journal":{"name":"International Journal of Computer Assisted Radiology and Surgery","volume":" ","pages":"2177-2186"},"PeriodicalIF":2.3000,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Computer Assisted Radiology and Surgery","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s11548-024-03065-7","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/29 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Purpose: Manual annotations for training deep learning models in auto-segmentation are time-intensive. This study introduces a hybrid representation-enhanced sampling strategy that integrates both density and diversity criteria within an uncertainty-based Bayesian active learning (BAL) framework to reduce annotation efforts by selecting the most informative training samples.

Methods: The experiments are performed on two lower extremity datasets of MRI and CT images, focusing on the segmentation of the femur, pelvis, sacrum, quadriceps femoris, hamstrings, adductors, sartorius, and iliopsoas, utilizing a U-net-based BAL framework. Our method selects uncertain samples with high density and diversity for manual revision, optimizing for maximal similarity to unlabeled instances and minimal similarity to existing training data. We assess the accuracy and efficiency using dice and a proposed metric called reduced annotation cost (RAC), respectively. We further evaluate the impact of various acquisition rules on BAL performance and design an ablation study for effectiveness estimation.

Results: In MRI and CT datasets, our method was superior or comparable to existing ones, achieving a 0.8% dice and 1.0% RAC increase in CT (statistically significant), and a 0.8% dice and 1.1% RAC increase in MRI (not statistically significant) in volume-wise acquisition. Our ablation study indicates that combining density and diversity criteria enhances the efficiency of BAL in musculoskeletal segmentation compared to using either criterion alone.

Conclusion: Our sampling method is proven efficient in reducing annotation costs in image segmentation tasks. The combination of the proposed method and our BAL framework provides a semi-automatic way for efficient annotation of medical image datasets.

Abstract Image

贝叶斯主动学习下肢肌肉骨骼分割中的混合表示增强采样。
目的:为训练自动分割中的深度学习模型而进行人工标注耗费大量时间。本研究介绍了一种混合表示增强采样策略,该策略在基于不确定性的贝叶斯主动学习(BAL)框架内整合了密度和多样性标准,通过选择信息量最大的训练样本来减少标注工作:实验在两个下肢数据集的核磁共振成像和 CT 图像上进行,重点是股骨、骨盆、骶骨、股四头肌、腘绳肌、内收肌、腓肠肌和髂腰肌的分割,并利用基于 U 网的 BAL 框架。我们的方法选择具有高密度和多样性的不确定样本进行人工修正,优化与未标记实例的最大相似性和与现有训练数据的最小相似性。我们分别使用骰子和一种称为降低标注成本(RAC)的拟议指标来评估准确性和效率。我们还进一步评估了各种采集规则对 BAL 性能的影响,并设计了一项消融研究来估算有效性:在 MRI 和 CT 数据集中,我们的方法优于或媲美现有的方法,在 CT 中实现了 0.8% 的骰子增加和 1.0% 的 RAC 增加(有统计学意义),在 MRI 中实现了 0.8% 的骰子增加和 1.1% 的 RAC 增加(无统计学意义)。我们的消融研究表明,与单独使用其中一种标准相比,结合密度和多样性标准可提高 BAL 在肌肉骨骼分割中的效率:结论:事实证明,我们的采样方法能有效降低图像分割任务中的注释成本。建议的方法与我们的 BAL 框架相结合,为高效注释医学图像数据集提供了一种半自动方法。
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
International Journal of Computer Assisted Radiology and Surgery
International Journal of Computer Assisted Radiology and Surgery ENGINEERING, BIOMEDICAL-RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
CiteScore
5.90
自引率
6.70%
发文量
243
审稿时长
6-12 weeks
期刊介绍: The International Journal for Computer Assisted Radiology and Surgery (IJCARS) is a peer-reviewed journal that provides a platform for closing the gap between medical and technical disciplines, and encourages interdisciplinary research and development activities in an international environment.
×
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