Seo Seung Yeon, Kim Soo-Jong, Oh Jungsu S, Chung Jinwha, Kim Seog-Young, Oh Seung Jun, Joo Segyeong, Kim Jae Seung
Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, Songpa-gu, South Korea.
Department of Biomedical Engineering, Asan Medical Center, University of Ulsan College of Medicine, Songpa-gu, South Korea.
Front Aging Neurosci. 2022 Mar 4;14:807903. doi: 10.3389/fnagi.2022.807903. eCollection 2022.
Although skull-stripping and brain region segmentation are essential for precise quantitative analysis of positron emission tomography (PET) of mouse brains, deep learning (DL)-based unified solutions, particularly for spatial normalization (SN), have posed a challenging problem in DL-based image processing. In this study, we propose an approach based on DL to resolve these issues. We generated both skull-stripping masks and individual brain-specific volumes-of-interest (VOIs-cortex, hippocampus, striatum, thalamus, and cerebellum) based on inverse spatial normalization (iSN) and deep convolutional neural network (deep CNN) models. We applied the proposed methods to mutated amyloid precursor protein and presenilin-1 mouse model of Alzheimer's disease. Eighteen mice underwent T2-weighted MRI and F FDG PET scans two times, before and after the administration of human immunoglobulin or antibody-based treatments. For training the CNN, manually traced brain masks and iSN-based target VOIs were used as the label. We compared our CNN-based VOIs with conventional (template-based) VOIs in terms of the correlation of standardized uptake value ratio (SUVR) by both methods and two-sample -tests of SUVR % changes in target VOIs before and after treatment. Our deep CNN-based method successfully generated brain parenchyma mask and target VOIs, which shows no significant difference from conventional VOI methods in SUVR correlation analysis, thus establishing methods of template-based VOI without SN.
尽管颅骨剥离和脑区分割对于小鼠脑正电子发射断层扫描（PET）的精确定量分析至关重要，但基于深度学习（DL）的统一解决方案，特别是用于空间归一化（SN）的方案，在基于DL的图像处理中提出了一个具有挑战性的问题。在本研究中，我们提出了一种基于DL的方法来解决这些问题。我们基于逆空间归一化（iSN）和深度卷积神经网络（深度CNN）模型生成了颅骨剥离掩码和个体脑特异性感兴趣体积（VOI-皮质、海马体、纹状体、丘脑和小脑）。我们将所提出的方法应用于阿尔茨海默病的突变淀粉样前体蛋白和早老素-1小鼠模型。18只小鼠在给予人免疫球蛋白或基于抗体的治疗之前和之后接受了两次T2加权MRI和F FDG PET扫描。为了训练CNN，将手动追踪的脑掩码和基于iSN的目标VOI用作标签。我们在两种方法的标准化摄取值比率（SUVR）的相关性以及治疗前后目标VOI中SUVR变化百分比的双样本检验方面，将基于CNN的VOI与传统（基于模板）的VOI进行了比较。我们基于深度CNN的方法成功生成了脑实质掩码和目标VOI，在SUVR相关性分析中与传统VOI方法没有显著差异，从而建立了无需SN的基于模板的VOI方法。
