Multi-perspective label based deep learning framework for cerebral vasculature segmentation in whole-brain fluorescence images

被引:3
作者
LI, Yuxin [1 ]
Ren, Tong [1 ]
LI, Junhuai [1 ]
LI, Xiangning [2 ,3 ]
LI, Anan [2 ,3 ]
机构
[1] Xian Univ Technol, Sch Comp Sci & Engn, Shaanxi Key Lab Network Comp & Secur Technol, Xian 710048, Peoples R China
[2] Huazhong Univ Sci & Technol, Sch Engn Sci, Britton Chance Ctr Biomed Photon, Wuhan Natl Lab Optoelect,MoE Key Lab Biomed Photo, Wuhan 430074, Peoples R China
[3] HUST, Suzhou Inst Brainsmat, Suzhou 215123, Peoples R China
基金
中国国家自然科学基金;
关键词
VESSEL SEGMENTATION; ENHANCEMENT; SYMMETRY;
D O I
10.1364/BOE.458111
中图分类号
Q5 [生物化学];
学科分类号
071010 ; 081704 ;
摘要
The popularity of fluorescent labelling and mesoscopic optical imaging techniques enable the acquisition of whole mammalian brain vasculature images at capillary resolution. Segmentation of the cerebrovascular network is essential for analyzing the cerebrovascular structure and revealing the pathogenesis of brain diseases. Existing deep learning methods use a single type of annotated labels with the same pixel weight to train the neural network and segment vessels. Due to the variation in the shape, density and brightness of vessels in whole-brain fluorescence images, it is difficult for a neural network trained with a single type of label to segment all vessels accurately. To address this problem, we proposed a deep learning cerebral vasculature segmentation framework based on multi-perspective labels. First, the pixels in the central region of thick vessels and the skeleton region of vessels were extracted separately using morphological operations based on the binary annotated labels to generate two different labels. Then, we designed a three-stage 3D convolutional neural network containing three sub-networks, namely thick-vessel enhancement network, vessel skeleton enhancement network and multi-channel fusion segmentation network. The first two sub-networks were trained by the two labels generated in the previous step, respectively, and pre-segmented the vessels. The third sub-network was responsible for fusing the pre-segmented results to precisely segment the vessels. We validated our method on two mouse cerebral vascular datasets generated by different fluorescence imaging modalities. The results showed that our method outperforms the state-of-the-art methods, and the proposed method can be applied to segment the vasculature on large-scale volumes.
引用
收藏
页码:3657 / 3671
页数:15
相关论文
共 40 条
[1]   Neuronal and Vascular Interactions [J].
Andreone, Benjamin J. ;
Lacoste, Baptiste ;
Gu, Chenghua .
ANNUAL REVIEW OF NEUROSCIENCE, VOL 38, 2015, 38 :25-46
[2]   Accurate Vessel Segmentation With Constrained B-Snake [J].
Cheng, Yuanzhi ;
Hu, Xin ;
Wang, Ji ;
Wang, Yadong ;
Tamura, Shinichi .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2015, 24 (08) :2440-2455
[3]  
Cicek Ozgun, 2016, Medical Image Computing and Computer-Assisted Intervention - MICCAI 2016. 19th International Conference. Proceedings: LNCS 9901, P424, DOI 10.1007/978-3-319-46723-8_49
[4]   Automatic Graph-Based Modeling of Brain Microvessels Captured With Two-Photon Microscopy [J].
Damseh, Rafat ;
Pouliot, Philippe ;
Gagnon, Louis ;
Sakadzic, Sava ;
Boas, David ;
Cheriet, Farida ;
Lesage, Frederic .
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2019, 23 (06) :2551-2562
[5]  
Frangi AF, 1998, LECT NOTES COMPUT SC, V1496, P130, DOI 10.1007/BFb0056195
[6]   High-throughput dual-colour precision imaging for brain-wide connectome with cytoarchitectonic landmarks at the cellular level [J].
Gong, Hui ;
Xu, Dongli ;
Yuan, Jing ;
Li, Xiangning ;
Guo, Congdi ;
Peng, Jie ;
Li, Yuxin ;
Schwarz, Lindsay A. ;
Li, Anan ;
Hu, Bihe ;
Xiong, Benyi ;
Sun, Qingtao ;
Zhang, Yalun ;
Liu, Jiepeng ;
Zhong, Qiuyuan ;
Xu, Tonghui ;
Zeng, Shaoqun ;
Luo, Qingming .
NATURE COMMUNICATIONS, 2016, 7
[7]   Model-Based Vasculature Extraction From Optical Fluorescence Cryomicrotome Images [J].
Goyal, Ayush ;
Lee, Jack ;
Lamata, Pablo ;
van den Wijngaard, Jeroen ;
van Horssen, Pepijn ;
Spaan, Jos ;
Siebes, Maria ;
Grau, Vicente ;
Smith, Nic P. .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2013, 32 (01) :56-72
[8]   Deep convolutional neural networks for segmenting 3D in vivo multiphoton images of vasculature in Alzheimer disease mouse models [J].
Haft-Javaherian, Mohammad ;
Fang, Linjing ;
Muse, Victorine ;
Schaffer, Chris B. ;
Nishimura, Nozomi ;
Sabuncu, Mert R. .
PLOS ONE, 2019, 14 (03)
[9]   Brain microvasculature has a common topology with local differences in geometry that match metabolic load [J].
Ji, Xiang ;
Ferreira, Tiago ;
Friedman, Beth ;
Liu, Rui ;
Liechty, Hannah ;
Bas, Erhan ;
Chandrashekar, Jayaram ;
Kleinfeld, David .
NEURON, 2021, 109 (07) :1168-1187.e13
[10]   Learning-based algorithms for vessel tracking: A review [J].
Jia, Dengqiang ;
Zhuang, Xiahai .
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2021, 89