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

Shuihua Wang; Yongyan Jiang; Xiaoxia Hou; Hong Cheng; Sidan Du

doi:10.1109/ACCESS.2017.2736558

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

Shuihua Wang^*, Yongyan Jiang, Xiaoxia Hou, Hong Cheng, Sidan Du

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

82 Citations (Scopus)

Abstract

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%.

Original language	English
Article number	8013653
Pages (from-to)	16576-16583
Number of pages	8
Journal	IEEE Access
Volume	5
DOIs	https://doi.org/10.1109/ACCESS.2017.2736558
Publication status	Published - 19 Aug 2017
Externally published	Yes

Keywords

Convolutional neural network
cerebral micro-bleed
network structure
rank based average pooling

Access to Document

10.1109/ACCESS.2017.2736558

Cite this

Wang, S., Jiang, Y., Hou, X., Cheng, H., & Du, S. (2017). Cerebral Micro-Bleed Detection Based on the Convolution Neural Network with Rank Based Average Pooling. IEEE Access, 5, 16576-16583. Article 8013653. https://doi.org/10.1109/ACCESS.2017.2736558

@article{aeeb3deae171490fb038f3137cde7c6f,

title = "Cerebral Micro-Bleed Detection Based on the Convolution Neural Network with Rank Based Average Pooling",

abstract = "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%.",

keywords = "Convolutional neural network, cerebral micro-bleed, network structure, rank based average pooling",

author = "Shuihua Wang and Yongyan Jiang and Xiaoxia Hou and Hong Cheng and Sidan Du",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2017",

month = aug,

day = "19",

doi = "10.1109/ACCESS.2017.2736558",

language = "English",

volume = "5",

pages = "16576--16583",

journal = "IEEE Access",

issn = "2169-3536",

}

TY - JOUR

T1 - Cerebral Micro-Bleed Detection Based on the Convolution Neural Network with Rank Based Average Pooling

AU - Wang, Shuihua

AU - Jiang, Yongyan

AU - Hou, Xiaoxia

AU - Cheng, Hong

AU - Du, Sidan

PY - 2017/8/19

Y1 - 2017/8/19

N2 - 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%.

AB - 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%.

KW - Convolutional neural network

KW - cerebral micro-bleed

KW - network structure

KW - rank based average pooling

UR - http://www.scopus.com/inward/record.url?scp=85028508754&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2017.2736558

DO - 10.1109/ACCESS.2017.2736558

M3 - Article

AN - SCOPUS:85028508754

SN - 2169-3536

VL - 5

SP - 16576

EP - 16583

JO - IEEE Access

JF - IEEE Access

M1 - 8013653

ER -

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

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this