ScarNet:一种基于晚期钆增强图像的自动心肌疤痕定量的新基础模型。

IF 6.1 1区 医学 Q1 CARDIAC & CARDIOVASCULAR SYSTEMS
Neda Tavakoli, Amir Ali Rahsepar, Brandon C Benefield, Daming Shen, Santiago López-Tapia, Florian Schiffers, Jeffrey J Goldberger, Christine M Albert, Edwin Wu, Aggelos K Katsaggelos, Daniel C Lee, Daniel Kim
{"title":"ScarNet:一种基于晚期钆增强图像的自动心肌疤痕定量的新基础模型。","authors":"Neda Tavakoli, Amir Ali Rahsepar, Brandon C Benefield, Daming Shen, Santiago López-Tapia, Florian Schiffers, Jeffrey J Goldberger, Christine M Albert, Edwin Wu, Aggelos K Katsaggelos, Daniel C Lee, Daniel Kim","doi":"10.1016/j.jocmr.2025.101945","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Late Gadolinium Enhancement (LGE) imaging remains the gold standard for assessing myocardial fibrosis and scarring, with left ventricular (LV) LGE presence and extent serving as a predictor of major adverse cardiac events (MACE). Despite its clinical significance, LGE-based LV scar quantification is not used routinely due to the labor-intensive manual segmentation and substantial inter-observer variability.</p><p><strong>Methods: </strong>We developed ScarNet that synergistically combines a transformer-based encoder in Medical Segment Anything Model (MedSAM), which we fine-tuned with our dataset, and a convolution-based decoder in U-Net with tailored attention blocks to automatically segment myocardial scar boundaries while maintaining anatomical context. This network was trained and fine-tuned on an existing database of 401 ischemic cardiomyopathy patients (4,137 2D LGE images) with expert segmentation of myocardial and scar boundaries in LGE images, validated on 100 patients (1,034 2D LGE images) during training, and tested on unseen set of 184 patients (1,895 2D LGE images). Ablation studies were conducted to validate each architectural component's contribution.</p><p><strong>Results: </strong>In 184 independent testing patients, ScarNet achieved accurate scar boundary segmentation (median DICE=0.912 [interquartile range (IQR): 0.863-0.944], concordance correlation coefficient [CCC]=0.963), significantly outperforming both MedSAM (median DICE=0.046 [IQR: 0.043-0.047], CCC=0.018) and nnU-Net (median DICE=0.638 [IQR: 0.604-0.661], CCC=0.734). For scar volume quantification, ScarNet demonstrated excellent agreement with manual analysis (CCC=0.995, percent bias=-0.63%, CoV=4.3%) compared to MedSAM (CCC=0.002, percent bias=-13.31%, CoV=130.3%) and nnU-Net (CCC=0.910, percent bias=-2.46%, CoV=20.3%). Similar trends were observed in the Monte Carlo simulations with noise perturbations. The overall accuracy was highest for SCARNet (sensitivity=95.3%; specificity=92.3%), followed by nnU-Net (sensitivity=74.9%; specificity=69.2%) and MedSAM (sensitivity=15.2%; specificity=92.3%).</p><p><strong>Conclusion: </strong>ScarNet outperformed MedSAM and nnU-Net for predicting myocardial and scar boundaries in LGE images of patients with ischemic cardiomyopathy. The Monte Carlo simulations demonstrated that ScarNet is less sensitive to noise perturbations than other tested networks.</p>","PeriodicalId":15221,"journal":{"name":"Journal of Cardiovascular Magnetic Resonance","volume":" ","pages":"101945"},"PeriodicalIF":6.1000,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"ScarNet: A Novel Foundation Model for Automated Myocardial Scar Quantification from Late Gadolinium-Enhancement Images.\",\"authors\":\"Neda Tavakoli, Amir Ali Rahsepar, Brandon C Benefield, Daming Shen, Santiago López-Tapia, Florian Schiffers, Jeffrey J Goldberger, Christine M Albert, Edwin Wu, Aggelos K Katsaggelos, Daniel C Lee, Daniel Kim\",\"doi\":\"10.1016/j.jocmr.2025.101945\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Late Gadolinium Enhancement (LGE) imaging remains the gold standard for assessing myocardial fibrosis and scarring, with left ventricular (LV) LGE presence and extent serving as a predictor of major adverse cardiac events (MACE). Despite its clinical significance, LGE-based LV scar quantification is not used routinely due to the labor-intensive manual segmentation and substantial inter-observer variability.</p><p><strong>Methods: </strong>We developed ScarNet that synergistically combines a transformer-based encoder in Medical Segment Anything Model (MedSAM), which we fine-tuned with our dataset, and a convolution-based decoder in U-Net with tailored attention blocks to automatically segment myocardial scar boundaries while maintaining anatomical context. This network was trained and fine-tuned on an existing database of 401 ischemic cardiomyopathy patients (4,137 2D LGE images) with expert segmentation of myocardial and scar boundaries in LGE images, validated on 100 patients (1,034 2D LGE images) during training, and tested on unseen set of 184 patients (1,895 2D LGE images). Ablation studies were conducted to validate each architectural component's contribution.</p><p><strong>Results: </strong>In 184 independent testing patients, ScarNet achieved accurate scar boundary segmentation (median DICE=0.912 [interquartile range (IQR): 0.863-0.944], concordance correlation coefficient [CCC]=0.963), significantly outperforming both MedSAM (median DICE=0.046 [IQR: 0.043-0.047], CCC=0.018) and nnU-Net (median DICE=0.638 [IQR: 0.604-0.661], CCC=0.734). For scar volume quantification, ScarNet demonstrated excellent agreement with manual analysis (CCC=0.995, percent bias=-0.63%, CoV=4.3%) compared to MedSAM (CCC=0.002, percent bias=-13.31%, CoV=130.3%) and nnU-Net (CCC=0.910, percent bias=-2.46%, CoV=20.3%). Similar trends were observed in the Monte Carlo simulations with noise perturbations. The overall accuracy was highest for SCARNet (sensitivity=95.3%; specificity=92.3%), followed by nnU-Net (sensitivity=74.9%; specificity=69.2%) and MedSAM (sensitivity=15.2%; specificity=92.3%).</p><p><strong>Conclusion: </strong>ScarNet outperformed MedSAM and nnU-Net for predicting myocardial and scar boundaries in LGE images of patients with ischemic cardiomyopathy. The Monte Carlo simulations demonstrated that ScarNet is less sensitive to noise perturbations than other tested networks.</p>\",\"PeriodicalId\":15221,\"journal\":{\"name\":\"Journal of Cardiovascular Magnetic Resonance\",\"volume\":\" \",\"pages\":\"101945\"},\"PeriodicalIF\":6.1000,\"publicationDate\":\"2025-08-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Cardiovascular Magnetic Resonance\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1016/j.jocmr.2025.101945\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CARDIAC & CARDIOVASCULAR SYSTEMS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Cardiovascular Magnetic Resonance","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1016/j.jocmr.2025.101945","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CARDIAC & CARDIOVASCULAR SYSTEMS","Score":null,"Total":0}
引用次数: 0

摘要

背景:晚期钆增强(LGE)成像仍然是评估心肌纤维化和瘢痕形成的金标准,左心室(LV) LGE的存在和程度可作为主要不良心脏事件(MACE)的预测指标。尽管具有临床意义,但基于lge的左室疤痕量化并没有被常规使用,因为它需要大量的人工分割和大量的观察者之间的差异。方法:我们开发了ScarNet,它协同结合了医疗片段任意模型(MedSAM)中基于变压器的编码器(我们根据我们的数据集进行了微调)和U-Net中基于卷积的解码器,该解码器具有定制的注意力块,可以在保持解剖背景的同时自动分割心肌疤痕边界。该网络在401例缺血性心肌病患者(4137张二维LGE图像)的现有数据库上进行训练和微调,对LGE图像中的心肌和疤痕边界进行专家分割,在训练期间对100例患者(1034张二维LGE图像)进行验证,并在未见的184例患者(1895张二维LGE图像)进行测试。进行消融研究以验证每个建筑组件的贡献。结果:在184例独立测试患者中,ScarNet实现了准确的疤痕边界分割(中位DICE=0.912[四分位间距(IQR): 0.863-0.944],一致性相关系数[CCC]=0.963),显著优于MedSAM(中位DICE=0.046 [IQR: 0.043-0.047], CCC=0.018)和nnU-Net(中位DICE=0.638 [IQR: 0.604-0.661], CCC=0.734)。对于疤痕体积定量,与MedSAM (CCC=0.002,百分比偏差=-13.31%,CoV=130.3%)和nnU-Net (CCC=0.910,百分比偏差=-2.46%,CoV=20.3%)相比,ScarNet与人工分析(CCC=0.995,百分比偏差=-0.63%,CoV=4.3%)表现出极好的一致性。在有噪声扰动的蒙特卡罗模拟中也观察到类似的趋势。SCARNet的总体准确性最高(灵敏度=95.3%,特异性=92.3%),其次是nnU-Net(灵敏度=74.9%,特异性=69.2%)和MedSAM(灵敏度=15.2%,特异性=92.3%)。结论:ScarNet在预测缺血性心肌病患者LGE图像的心肌和瘢痕边界方面优于MedSAM和nnU-Net。蒙特卡罗仿真表明,ScarNet对噪声干扰的敏感性低于其他测试网络。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
ScarNet: A Novel Foundation Model for Automated Myocardial Scar Quantification from Late Gadolinium-Enhancement Images.

Background: Late Gadolinium Enhancement (LGE) imaging remains the gold standard for assessing myocardial fibrosis and scarring, with left ventricular (LV) LGE presence and extent serving as a predictor of major adverse cardiac events (MACE). Despite its clinical significance, LGE-based LV scar quantification is not used routinely due to the labor-intensive manual segmentation and substantial inter-observer variability.

Methods: We developed ScarNet that synergistically combines a transformer-based encoder in Medical Segment Anything Model (MedSAM), which we fine-tuned with our dataset, and a convolution-based decoder in U-Net with tailored attention blocks to automatically segment myocardial scar boundaries while maintaining anatomical context. This network was trained and fine-tuned on an existing database of 401 ischemic cardiomyopathy patients (4,137 2D LGE images) with expert segmentation of myocardial and scar boundaries in LGE images, validated on 100 patients (1,034 2D LGE images) during training, and tested on unseen set of 184 patients (1,895 2D LGE images). Ablation studies were conducted to validate each architectural component's contribution.

Results: In 184 independent testing patients, ScarNet achieved accurate scar boundary segmentation (median DICE=0.912 [interquartile range (IQR): 0.863-0.944], concordance correlation coefficient [CCC]=0.963), significantly outperforming both MedSAM (median DICE=0.046 [IQR: 0.043-0.047], CCC=0.018) and nnU-Net (median DICE=0.638 [IQR: 0.604-0.661], CCC=0.734). For scar volume quantification, ScarNet demonstrated excellent agreement with manual analysis (CCC=0.995, percent bias=-0.63%, CoV=4.3%) compared to MedSAM (CCC=0.002, percent bias=-13.31%, CoV=130.3%) and nnU-Net (CCC=0.910, percent bias=-2.46%, CoV=20.3%). Similar trends were observed in the Monte Carlo simulations with noise perturbations. The overall accuracy was highest for SCARNet (sensitivity=95.3%; specificity=92.3%), followed by nnU-Net (sensitivity=74.9%; specificity=69.2%) and MedSAM (sensitivity=15.2%; specificity=92.3%).

Conclusion: ScarNet outperformed MedSAM and nnU-Net for predicting myocardial and scar boundaries in LGE images of patients with ischemic cardiomyopathy. The Monte Carlo simulations demonstrated that ScarNet is less sensitive to noise perturbations than other tested networks.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
10.90
自引率
12.50%
发文量
61
审稿时长
6-12 weeks
期刊介绍: Journal of Cardiovascular Magnetic Resonance (JCMR) publishes high-quality articles on all aspects of basic, translational and clinical research on the design, development, manufacture, and evaluation of cardiovascular magnetic resonance (CMR) methods applied to the cardiovascular system. Topical areas include, but are not limited to: New applications of magnetic resonance to improve the diagnostic strategies, risk stratification, characterization and management of diseases affecting the cardiovascular system. New methods to enhance or accelerate image acquisition and data analysis. Results of multicenter, or larger single-center studies that provide insight into the utility of CMR. Basic biological perceptions derived by CMR methods.
×
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