Pyramid Predictive Attention Network for Medical Image Segmentation

Tingxiao YANG; Yuichiro YOSHIMURA; Akira MORITA; Takao NAMIKI; Toshiya NAKAGUCHI

doi:10.1587/transfun.E102.A.1225

Open Access
Pyramid Predictive Attention Network for Medical Image Segmentation

Tingxiao YANG, Yuichiro YOSHIMURA, Akira MORITA, Takao NAMIKI, Toshiya NAKAGUCHI

Full Text Views

102

Cite this

Free PDF (6MB)

Summary :

In this paper, we propose a Pyramid Predictive Attention Network (PPAN) for medical image segmentation. In the medical field, the size of dataset generally restricts the performance of deep CNN and deploying the trained network with gross parameters into the terminal device with limited memory is an expectation. Our team aims to the future home medical diagnosis and search for lightweight medical image segmentation network. Therefore, we designed PPAN mainly made of Xception blocks which are modified from DeepLab v3+ and consist of separable depthwise convolutions to speed up the computation and reduce the parameters. Meanwhile, by utilizing pyramid predictions from each dimension stage will guide the network more accessible to optimize the training process towards the final segmentation target without degrading the performance. IoU metric is used for the evaluation on the test dataset. We compared our designed network performance with the current state of the art segmentation networks on our RGB tongue dataset which was captured by the developed TIAS system for tongue diagnosis. Our designed network reduced 80 percentage parameters compared to the most widely used U-Net in medical image segmentation and achieved similar or better performance. Any terminal with limited storage which is needed a segment of RGB image can refer to our designed PPAN.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E102-A No.9 pp.1225-1234

Publication Date: 2019/09/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E102.A.1225

Type of Manuscript: Special Section PAPER (Special Section on Image Media Quality)

Category

Authors

Tingxiao YANG
  Chiba University
Yuichiro YOSHIMURA
  Chiba University
Akira MORITA
  Chiba University
Takao NAMIKI
  Chiba University
Toshiya NAKAGUCHI
  Chiba University

Keyword

PPAN, CNN, predictive, separable, IoU, segmentation, medical

Cite this

Copy

Tingxiao YANG, Yuichiro YOSHIMURA, Akira MORITA, Takao NAMIKI, Toshiya NAKAGUCHI, "Pyramid Predictive Attention Network for Medical Image Segmentation" in IEICE TRANSACTIONS on Fundamentals, vol. E102-A, no. 9, pp. 1225-1234, September 2019, doi: 10.1587/transfun.E102.A.1225.
Abstract: In this paper, we propose a Pyramid Predictive Attention Network (PPAN) for medical image segmentation. In the medical field, the size of dataset generally restricts the performance of deep CNN and deploying the trained network with gross parameters into the terminal device with limited memory is an expectation. Our team aims to the future home medical diagnosis and search for lightweight medical image segmentation network. Therefore, we designed PPAN mainly made of Xception blocks which are modified from DeepLab v3+ and consist of separable depthwise convolutions to speed up the computation and reduce the parameters. Meanwhile, by utilizing pyramid predictions from each dimension stage will guide the network more accessible to optimize the training process towards the final segmentation target without degrading the performance. IoU metric is used for the evaluation on the test dataset. We compared our designed network performance with the current state of the art segmentation networks on our RGB tongue dataset which was captured by the developed TIAS system for tongue diagnosis. Our designed network reduced 80 percentage parameters compared to the most widely used U-Net in medical image segmentation and achieved similar or better performance. Any terminal with limited storage which is needed a segment of RGB image can refer to our designed PPAN.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E102.A.1225/_p

Copy

@ARTICLE{e102-a_9_1225,
author={Tingxiao YANG, Yuichiro YOSHIMURA, Akira MORITA, Takao NAMIKI, Toshiya NAKAGUCHI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Pyramid Predictive Attention Network for Medical Image Segmentation},
year={2019},
volume={E102-A},
number={9},
pages={1225-1234},
abstract={In this paper, we propose a Pyramid Predictive Attention Network (PPAN) for medical image segmentation. In the medical field, the size of dataset generally restricts the performance of deep CNN and deploying the trained network with gross parameters into the terminal device with limited memory is an expectation. Our team aims to the future home medical diagnosis and search for lightweight medical image segmentation network. Therefore, we designed PPAN mainly made of Xception blocks which are modified from DeepLab v3+ and consist of separable depthwise convolutions to speed up the computation and reduce the parameters. Meanwhile, by utilizing pyramid predictions from each dimension stage will guide the network more accessible to optimize the training process towards the final segmentation target without degrading the performance. IoU metric is used for the evaluation on the test dataset. We compared our designed network performance with the current state of the art segmentation networks on our RGB tongue dataset which was captured by the developed TIAS system for tongue diagnosis. Our designed network reduced 80 percentage parameters compared to the most widely used U-Net in medical image segmentation and achieved similar or better performance. Any terminal with limited storage which is needed a segment of RGB image can refer to our designed PPAN.},
keywords={},
doi={10.1587/transfun.E102.A.1225},
ISSN={1745-1337},
month={September},}

Copy

TY - JOUR
TI - Pyramid Predictive Attention Network for Medical Image Segmentation
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1225
EP - 1234
AU - Tingxiao YANG
AU - Yuichiro YOSHIMURA
AU - Akira MORITA
AU - Takao NAMIKI
AU - Toshiya NAKAGUCHI
PY - 2019
DO - 10.1587/transfun.E102.A.1225
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E102-A
IS - 9
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - September 2019
AB - In this paper, we propose a Pyramid Predictive Attention Network (PPAN) for medical image segmentation. In the medical field, the size of dataset generally restricts the performance of deep CNN and deploying the trained network with gross parameters into the terminal device with limited memory is an expectation. Our team aims to the future home medical diagnosis and search for lightweight medical image segmentation network. Therefore, we designed PPAN mainly made of Xception blocks which are modified from DeepLab v3+ and consist of separable depthwise convolutions to speed up the computation and reduce the parameters. Meanwhile, by utilizing pyramid predictions from each dimension stage will guide the network more accessible to optimize the training process towards the final segmentation target without degrading the performance. IoU metric is used for the evaluation on the test dataset. We compared our designed network performance with the current state of the art segmentation networks on our RGB tongue dataset which was captured by the developed TIAS system for tongue diagnosis. Our designed network reduced 80 percentage parameters compared to the most widely used U-Net in medical image segmentation and achieved similar or better performance. Any terminal with limited storage which is needed a segment of RGB image can refer to our designed PPAN.
ER -