Cerebral Micro-Bleed Detection Based on the Convolution Neural Network With Rank Based Average Pooling

被引：76

作者：

Wang, Shuihua ^{[1
]}

Jiang, Yongyan ^{[2
]}

Hou, Xiaoxia ^{[3
]}

Cheng, Hong ^{[3
]}

Dui, Sidan ^{[1
]}

机构：

[1] Nanjing Univ, Sch Elect Engn, Nanjing 210046, Jiangsu, Peoples R China

[2] Zhongyuan Univ Technol, Coll Sci, Zhengzhou 450007, Henan, Peoples R China

[3] Nanjing Med Univ, Dept Neurol, Affiliated Hosp 1, Nanjing 210029, Jiangsu, Peoples R China

来源：

IEEE ACCESS | 2017年 / 5卷

关键词：

Convolutional neural network; cerebral micro-bleed; network structure; rank based average; pooling; COMPUTER-AIDED DETECTION; MICROBLEEDS;

D O I：

10.1109/ACCESS.2017.2736558

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Cerebral micro-bleed (CMB) is small perivascular hemosiderin deposits from leakage through cerebral small vessels. They can result from cerebra-vascular disease, dementia, or simply from normal aging. It can be visualized via the susceptibility weighted imaging (SWI). Based on the SWI, we propose to use different structures of the CNN with rank-based average pooling to detect the CMB, and compare this method used in this paper to the current state-of-the-art methods. We can find that the CNN with five layers obtains the best performance, with a sensitivity of 96.94%, a specificity of 97.18%, and an accuracy of 97.18%.

引用

页码：16576 / 16583

页数：8

共 21 条

[11] Rank Pooling for Action Recognition [J].

Fernando, Basura ;

Gavves, Efstratios ;

Oramas, Jose M. ;

Ghodrati, Amir ;

Tuytelaars, Tinne .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2017, 39 (04) :773-787

[12]

Giusti A, 2013, IEEE IMAGE PROC, P4034, DOI 10.1109/ICIP.2013.6738831

[13] Digital selection and analogue amplification coexist in a cortex-inspired silicon circuit (vol 405, pg 947, 2000) [J].

Hahnloser, RHR ;

Sarpeshkar, R ;

Mahowald, MA ;

Douglas, RJ ;

Seung, S .

NATURE, 2000, 408 (6815) :1012-U24

[14] Efficient detection of cerebral microbleeds on 7.0 T MR images using the radial symmetry transform [J].

Kuijf, Hugo J. ;

de Bresser, Jeroen ;

Geerlings, Mirjam I. ;

Conijn, Mandy M. A. ;

Viergever, Max A. ;

Biessels, Geert Jan ;

Vincken, Koen L. .

NEUROIMAGE, 2012, 59 (03) :2266-2273

[15] Gradient-based learning applied to document recognition [J].

Lecun, Y ;

Bottou, L ;

Bengio, Y ;

Haffner, P .

PROCEEDINGS OF THE IEEE, 1998, 86 (11) :2278-2324

[16] Visual Saliency Detection Based on Multiscale Deep CNN Features [J].

Li, Guanbin ;

Yu, Yizhou .

IEEE TRANSACTIONS ON IMAGE PROCESSING, 2016, 25 (11) :5012-5024

[17] Affinity CNN: Learning Pixel-Centric Pairwise Relations for Figure/Ground Embedding [J].

Maire, Michael ;

Narihira, Takuya ;

Yu, Stella X. .

2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :174-182

[18] Rank-based pooling for deep convolutional neural networks [J].

Shi, Zenglin ;

Ye, Yangdong ;

Wu, Yunpeng .

NEURAL NETWORKS, 2016, 83 :21-31

[19] Automated detection of cerebral microbleeds in patients with traumatic brain injury [J].

van den Heuvel, T. L. A. ;

van der Eerden, A. W. ;

Manniesing, R. ;

Ghafoorian, M. ;

Tan, T. ;

Andriessen, T. M. J. C. ;

Vyvere, T. Vande ;

van den Hauwe, L. ;

Romeny, B. M. ter Haar ;

Goraj, B. M. ;

Platel, B. .

NEUROIMAGE-CLINICAL, 2016, 12 :241-251

[20] Three-stream CNNs for action recognition [J].

Wang, Liangliang ;

Ge, Lianzheng ;

Li, Ruifeng ;

Fang, Yajun .

PATTERN RECOGNITION LETTERS, 2017, 92 :33-40

← 1 2 3 →