Semi-supervised Learning for Generalizable Intracranial Hemorrhage Detection and Segmentation.

IF 8.1 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE Radiology-Artificial Intelligence Pub Date : 2024-05-01 DOI:10.1148/ryai.230077
Emily Lin, Esther L Yuh
{"title":"Semi-supervised Learning for Generalizable Intracranial Hemorrhage Detection and Segmentation.","authors":"Emily Lin, Esther L Yuh","doi":"10.1148/ryai.230077","DOIUrl":null,"url":null,"abstract":"<p><p>Purpose To develop and evaluate a semi-supervised learning model for intracranial hemorrhage detection and segmentation on an out-of-distribution head CT evaluation set. Materials and Methods This retrospective study used semi-supervised learning to bootstrap performance. An initial \"teacher\" deep learning model was trained on 457 pixel-labeled head CT scans collected from one U.S. institution from 2010 to 2017 and used to generate pseudo labels on a separate unlabeled corpus of 25 000 examinations from the Radiological Society of North America and American Society of Neuroradiology. A second \"student\" model was trained on this combined pixel- and pseudo-labeled dataset. Hyperparameter tuning was performed on a validation set of 93 scans. Testing for both classification (<i>n</i> = 481 examinations) and segmentation (<i>n</i> = 23 examinations, or 529 images) was performed on CQ500, a dataset of 481 scans performed in India, to evaluate out-of-distribution generalizability. The semi-supervised model was compared with a baseline model trained on only labeled data using area under the receiver operating characteristic curve, Dice similarity coefficient, and average precision metrics. Results The semi-supervised model achieved a statistically significant higher examination area under the receiver operating characteristic curve on CQ500 compared with the baseline (0.939 [95% CI: 0.938, 0.940] vs 0.907 [95% CI: 0.906, 0.908]; <i>P</i> = .009). It also achieved a higher Dice similarity coefficient (0.829 [95% CI: 0.825, 0.833] vs 0.809 [95% CI: 0.803, 0.812]; <i>P</i> = .012) and pixel average precision (0.848 [95% CI: 0.843, 0.853]) vs 0.828 [95% CI: 0.817, 0.828]) compared with the baseline. Conclusion The addition of unlabeled data in a semi-supervised learning framework demonstrates stronger generalizability potential for intracranial hemorrhage detection and segmentation compared with a supervised baseline. <b>Keywords:</b> Semi-supervised Learning, Traumatic Brain Injury, CT, Machine Learning <i>Supplemental material is available for this article.</i> Published under a CC BY 4.0 license. See also the commentary by Swimburne in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":8.1000,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11140498/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Radiology-Artificial Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1148/ryai.230077","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

Abstract

Purpose To develop and evaluate a semi-supervised learning model for intracranial hemorrhage detection and segmentation on an out-of-distribution head CT evaluation set. Materials and Methods This retrospective study used semi-supervised learning to bootstrap performance. An initial "teacher" deep learning model was trained on 457 pixel-labeled head CT scans collected from one U.S. institution from 2010 to 2017 and used to generate pseudo labels on a separate unlabeled corpus of 25 000 examinations from the Radiological Society of North America and American Society of Neuroradiology. A second "student" model was trained on this combined pixel- and pseudo-labeled dataset. Hyperparameter tuning was performed on a validation set of 93 scans. Testing for both classification (n = 481 examinations) and segmentation (n = 23 examinations, or 529 images) was performed on CQ500, a dataset of 481 scans performed in India, to evaluate out-of-distribution generalizability. The semi-supervised model was compared with a baseline model trained on only labeled data using area under the receiver operating characteristic curve, Dice similarity coefficient, and average precision metrics. Results The semi-supervised model achieved a statistically significant higher examination area under the receiver operating characteristic curve on CQ500 compared with the baseline (0.939 [95% CI: 0.938, 0.940] vs 0.907 [95% CI: 0.906, 0.908]; P = .009). It also achieved a higher Dice similarity coefficient (0.829 [95% CI: 0.825, 0.833] vs 0.809 [95% CI: 0.803, 0.812]; P = .012) and pixel average precision (0.848 [95% CI: 0.843, 0.853]) vs 0.828 [95% CI: 0.817, 0.828]) compared with the baseline. Conclusion The addition of unlabeled data in a semi-supervised learning framework demonstrates stronger generalizability potential for intracranial hemorrhage detection and segmentation compared with a supervised baseline. Keywords: Semi-supervised Learning, Traumatic Brain Injury, CT, Machine Learning Supplemental material is available for this article. Published under a CC BY 4.0 license. See also the commentary by Swimburne in this issue.

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
用于颅内出血检测和分割的半监督学习。
"刚刚接受 "的论文经过同行评审,已被接受在《放射学》上发表:人工智能》上发表。这篇文章在以最终版本发表之前,还将经过校对、排版和校对审核。请注意,在制作最终校对稿的过程中,可能会发现影响文章内容的错误。目的 在分布外头部 CT 评估集上开发和评估用于颅内出血检测和分割的半监督学习模型。材料与方法 这项回顾性研究使用半监督学习来引导性能。最初的 "教师 "深度学习模型是在 2010-2017 年间从一家美国机构收集的 457 个像素标记的头部 CT 扫描上训练的,并用于在来自 RSNA 和 ASNR 的 25,000 次检查的单独无标记语料库上生成伪标签。第二个 "学生 "模型是在这个像素与伪标签相结合的数据集上进行训练的。超参数调整在 93 个扫描的验证集上进行。分类(n = 481 次检查)和分割(n = 23 次检查,或 529 张图像)测试在 CQ500(印度进行的 481 次扫描的数据集)上进行,以评估分布外的通用性。使用接收者工作特征曲线下面积 (AUC)、Dice 相似性系数 (DSC) 和平均精确度 (AP) 指标,将半监督模型与仅在标记数据上训练的基线模型进行比较。结果 与基线模型相比,半监督模型在 CQ500 上的检查 AUC 明显更高(0.939 [0.938, 0.940] 对 0.907 [0.906, 0.908])(P = .009)。与基线相比,DSC(0.829 [0.825, 0.833] 对 0.809 [0.803, 0.812])(P = .012)和 Pixel AP(0.848 [0.843, 0.853])对 0.828 [0.817, 0.828])也更高。结论 与监督基线相比,在半监督学习框架中加入无标记数据,可为颅内出血检测和分割提供更强的通用性。©RSNA, 2024.
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
CiteScore
16.20
自引率
1.00%
发文量
0
期刊介绍: Radiology: Artificial Intelligence is a bi-monthly publication that focuses on the emerging applications of machine learning and artificial intelligence in the field of imaging across various disciplines. This journal is available online and accepts multiple manuscript types, including Original Research, Technical Developments, Data Resources, Review articles, Editorials, Letters to the Editor and Replies, Special Reports, and AI in Brief.
期刊最新文献
Integrated Deep Learning Model for the Detection, Segmentation, and Morphologic Analysis of Intracranial Aneurysms Using CT Angiography. RSNA 2023 Abdominal Trauma AI Challenge Review and Outcomes Analysis. SCIseg: Automatic Segmentation of Intramedullary Lesions in Spinal Cord Injury on T2-weighted MRI Scans. Combining Biology-based and MRI Data-driven Modeling to Predict Response to Neoadjuvant Chemotherapy in Patients with Triple-Negative Breast Cancer. Optimizing Performance of Transformer-based Models for Fetal Brain MR Image Segmentation.
×
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