{"title":"Brain Lesion Synthesis via Progressive Adversarial Variational Auto-Encoder.","authors":"Jiayu Huo, Vejay Vakharia, Chengyuan Wu, Ashwini Sharan, Andrew Ko, Sébastien Ourselin, Rachel Sparks","doi":"10.1007/978-3-031-16980-9_10","DOIUrl":null,"url":null,"abstract":"<p><p>Laser interstitial thermal therapy (LITT) is a novel minimally invasive treatment that is used to ablate intracranial structures to treat mesial temporal lobe epilepsy (MTLE). Region of interest (ROI) segmentation before and after LITT would enable automated lesion quantification to objectively assess treatment efficacy. Deep learning techniques, such as convolutional neural networks (CNNs) are state-of-the-art solutions for ROI segmentation, but require large amounts of annotated data during the training. However, collecting large datasets from emerging treatments such as LITT is impractical. In this paper, we propose a progressive brain lesion synthesis framework (PAVAE) to expand both the quantity and diversity of the training dataset. Concretely, our framework consists of two sequential networks: a mask synthesis network and a mask-guided lesion synthesis network. To better employ extrinsic information to provide additional supervision during network training, we design a condition embedding block (CEB) and a mask embedding block (MEB) to encode inherent conditions of masks to the feature space. Finally, a segmentation network is trained using raw and synthetic lesion images to evaluate the effectiveness of the proposed framework. Experimental results show that our method can achieve realistic synthetic results and boost the performance of down-stream segmentation tasks above traditional data augmentation techniques.</p>","PeriodicalId":91967,"journal":{"name":"Simulation and synthesis in medical imaging : ... International Workshop, SASHIMI ..., held in conjunction with MICCAI ..., proceedings. SASHIMI (Workshop)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7616255/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Simulation and synthesis in medical imaging : ... International Workshop, SASHIMI ..., held in conjunction with MICCAI ..., proceedings. SASHIMI (Workshop)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/978-3-031-16980-9_10","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Laser interstitial thermal therapy (LITT) is a novel minimally invasive treatment that is used to ablate intracranial structures to treat mesial temporal lobe epilepsy (MTLE). Region of interest (ROI) segmentation before and after LITT would enable automated lesion quantification to objectively assess treatment efficacy. Deep learning techniques, such as convolutional neural networks (CNNs) are state-of-the-art solutions for ROI segmentation, but require large amounts of annotated data during the training. However, collecting large datasets from emerging treatments such as LITT is impractical. In this paper, we propose a progressive brain lesion synthesis framework (PAVAE) to expand both the quantity and diversity of the training dataset. Concretely, our framework consists of two sequential networks: a mask synthesis network and a mask-guided lesion synthesis network. To better employ extrinsic information to provide additional supervision during network training, we design a condition embedding block (CEB) and a mask embedding block (MEB) to encode inherent conditions of masks to the feature space. Finally, a segmentation network is trained using raw and synthetic lesion images to evaluate the effectiveness of the proposed framework. Experimental results show that our method can achieve realistic synthetic results and boost the performance of down-stream segmentation tasks above traditional data augmentation techniques.
激光间质热疗(LITT)是一种新型微创疗法,用于消融颅内结构以治疗颞叶中叶癫痫(MTLE)。LITT 治疗前后的兴趣区域(ROI)分割可实现自动病变量化,从而客观评估治疗效果。卷积神经网络(CNN)等深度学习技术是最先进的 ROI 分割解决方案,但在训练过程中需要大量的注释数据。然而,从 LITT 等新兴疗法中收集大量数据集是不切实际的。在本文中,我们提出了一个渐进式脑病变合成框架(PAVAE),以扩大训练数据集的数量和多样性。具体来说,我们的框架由两个连续的网络组成:一个掩膜合成网络和一个掩膜引导的病变合成网络。为了在网络训练过程中更好地利用外在信息提供额外的监督,我们设计了一个条件嵌入块(CEB)和一个掩膜嵌入块(MEB),将掩膜的固有条件编码到特征空间中。最后,我们使用原始病变图像和合成病变图像对分割网络进行了训练,以评估所提出的框架的有效性。实验结果表明,我们的方法能获得逼真的合成结果,并能提升下流分割任务的性能,超过传统的数据增强技术。