arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.00166eess.IVcs.CVcs.LG
收藏

使用深度学习自动分割血管内OCT图像中的微血管

Automated segmentation of microvessels in intravascular OCT images using deep learning

Case Western Reserve University(凯斯西储大学);The Ohio State University(俄亥俄州立大学);The Hashemite University(哈希姆大学);University Hospitals Cleveland Medical Center(克利夫兰大学医院医疗中心);Galeazzi San’Ambrogio Hospital(加拉齐圣安布罗焦医院);University of South Florida(南佛罗里达大学)
Juhwan Lee, Justin N. Kim, Lia Gomez-Perez, Yazan Gharaibeh, Issam Motairek, Ga-briel T. R. Pereira, Vladislav N. Zimin, Luis A. P. Dallan, Ammar Hoori, Sadeer … 展开作者

Juhwan Lee, Justin N. Kim, Lia Gomez-Perez, Yazan Gharaibeh, Issam Motairek, Ga-briel T. R. Pereira, Vladislav N. Zimin, Luis A. P. Dallan, Ammar Hoori, Sadeer Al-Kindi, Giulio Guagliumi, Hiram G. Bezerra, David L. Wilson

摘要:

为了分析这种脆弱性,我们开发了一种自动深度学习方法,用于检测血管内光学相干断层扫描(IVOCT)图像中的微血管。共分析了85个病变和37个正常节段的8,403个IVOCT图像帧。使用我们小组以前开发的专用软件(OCTOPUS)进行手动注释。将极(r,θ)域中的数据增强应用于原始IVOCT图像,以确保微血管出现在所有可能的角度。预处理方法包括导丝/阴影检测、管腔分割、像素移位和降噪。使用DeepLab v3+分割微血管候选者。使用浅卷积神经网络将每个候选者上的边界框分类为微血管或非微血管。为了更好地分类,我们使用了数据增强(即,角度旋转)。数据增强和预处理步骤显著改善了微血管分割性能,产生了Dice为0.71+/-0.10和87.7+/-6.6%/99.8+/-0.1%像素灵敏度/特异性的方法。用于从候选人中分类微血管的网络表现非常好,灵敏度为99.5+/-0.3%,特异性为98.8+/-1.0%,准确性为99.1+/-0.5%。分类步骤消除了大多数残留假阳性,并且Dice系数从0.71增加到0.73。此外,我们的方法产生了698个存在微血管的图像帧,与手动分析的730个相比,差异为4.4%。与手动方法相比,自动方法改善了微血管连续性,这意味着分割性能得到改善。该方法将是有用的研究目的,以及潜在的未来治疗计划。

英文摘要:

To analyze this characteristic of vulnerability, we developed an automated deep learning method for detecting microvessels in intravascular optical coherence tomography (IVOCT) images. A total of 8,403 IVOCT image frames from 85 lesions and 37 normal segments were analyzed. Manual annotation was done using a dedicated software (OCTOPUS) previously developed by our group. Data augmentation in the polar (r,θ) domain was applied to raw IVOCT images to ensure that microvessels appear at all possible angles. Pre-processing methods included guidewire/shadow detection, lumen segmentation, pixel shifting, and noise reduction. DeepLab v3+ was used to segment microvessel candidates. A bounding box on each candidate was classified as either microvessel or non-microvessel using a shallow convolutional neural network. For better classification, we used data augmentation (i.e., angle rotation) on bounding boxes with a microvessel during network training. Data augmentation and pre-processing steps improved microvessel segmentation performance significantly, yielding a method with Dice of 0.71+/-0.10 and pixel-wise sensitivity/specificity of 87.7+/-6.6%/99.8+/-0.1%. The network for classifying microvessels from candidates performed exceptionally well, with sensitivity of 99.5+/-0.3%, specificity of 98.8+/-1.0%, and accuracy of 99.1+/-0.5%. The classification step eliminated the majority of residual false positives, and the Dice coefficient increased from 0.71 to 0.73. In addition, our method produced 698 image frames with microvessels present, compared to 730 from manual analysis, representing a 4.4% difference. When compared to the manual method, the automated method improved microvessel continuity, implying improved segmentation performance. The method will be useful for research purposes as well as potential future treatment planning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑