Capsules for biomedical image segmentation.

Capsules for biomedical image segmentation.
复制标题

用于生物医学图像分割的胶囊。

DOI:
10.1016/j.media.2020.101889
复制
发表时间:
2021-03
影响因子:
10.9
通讯作者:
Bagci U
Bagci U
中科院分区:
工程技术1区
文献类型:
--
作者:
LaLonde R;Xu Z;Irmakci I;Jain S;Bagci U

文献摘要

参考文献

被引文献

相似文献

我们的工作在文献中首次将胶囊网络的使用扩展到对象分割任务。这是通过引入局部约束路由和变换矩阵共享而实现的,这减少了参数/内存负担,并允许以大分辨率分割对象。为了补偿在约束路由过程中全局信息的丢失,我们提出了“解卷积”胶囊的概念来创建一个深度编码器-解码器风格的网络,称为SegCaps。我们将掩蔽重建正则化扩展到分割任务,并对我们方法的每个组件进行彻底的消融实验。所提出的卷积-去卷积胶囊网络SegCaps在使用流行分割网络的一小部分参数的同时显示了最先进的结果。为了验证我们提出的方法,我们进行实验分割病理肺临床和临床前胸部计算机断层扫描(CT)扫描和分割肌肉和脂肪(脂肪)组织的磁共振成像(MRI)扫描人类受试者的大腿。值得注意的是,我们在肺分割方面的实验代表了文献中病理肺分割方面最大规模的研究,我们在五个极具挑战性的数据集上进行了实验,这些数据集包含临床和临床前受试者以及近2000个计算机断层扫描。我们新开发的分割平台在所有数据集上都优于其他方法,同时利用流行的U-Net中不到5%的参数进行生物医学图像分割。此外,我们证明了胶囊的能力,推广到看不见的处理自然图像上的旋转/反射。
Our work expands the use of capsule networks to the task of object segmentation for the first time in the literature. This is made possible via the introduction of locally-constrained routing and transformation matrix sharing, which reduces the parameter/memory burden and allows for the segmentation of objects at large resolutions. To compensate for the loss of global information in constraining the routing, we propose the concept of “deconvolutional” capsules to create a deep encoder-decoder style network, called SegCaps. We extend the masked reconstruction regularization to the task of segmentation and perform thorough ablation experiments on each component of our method. The proposed convolutional-deconvolutional capsule network, SegCaps, shows state-of-the-art results while using a fraction of the parameters of popular segmentation networks. To validate our proposed method, we perform experiments segmenting pathological lungs from clinical and pre-clinical thoracic computed tomography (CT) scans and segmenting muscle and adipose (fat) tissue from magnetic resonance imaging (MRI) scans of human subjects’ thighs. Notably, our experiments in lung segmentation represent the largest-scale study in pathological lung segmentation in the literature, where we conduct experiments across five extremely challenging datasets, containing both clinical and pre-clinical subjects, and nearly 2000 computed-tomography scans. Our newly developed segmentation platform outperforms other methods across all datasets while utilizing less than 5% of the parameters in the popular U-Net for biomedical image segmentation. Further, we demonstrate capsules’ ability to generalize to unseen handling of rotations/reflections on natural images.
DOI: 10.1023/b:visi.0000022288.19776.77
发表时间: 2004-09-01
影响因子: 19.5
作者:
Felzenszwalb, PF;Huttenlocher, DP
通讯作者: Huttenlocher, DP
DOI: 10.1109/tmi.2011.2180920
发表时间: 2012-03-01
影响因子: 10.6
作者:
Bagci, Ulas;Chen, Xinjian;Udupa, Jayaram K.
通讯作者: Udupa, Jayaram K.
DOI: 10.1109/tpami.2006.233
发表时间: 2006-11-01
影响因子: 23.6
作者:
Grady, Leo
通讯作者: Grady, Leo
DOI: 10.1016/j.compmedimag.2011.07.003
发表时间: 2012-04-01
影响因子: 5.7
作者:
Depeursinge, Adrien;Vargas, Alejandro;Mueller, Henning
通讯作者: Mueller, Henning
DOI: 10.1007/978-1-4419-8204-9_1
发表时间: 2011-01-01
期刊: MULTI MODALITY STATE-OF-THE-ART MEDICAL IMAGE SEGMENTATION AND REGISTRATION METHODOLOGIES, VOL II
影响因子: --
作者:
Elnakib, Ahmed;Gimel'farb, Georgy;El-Baz, Ayman
通讯作者: El-Baz, Ayman