TY - JOUR
T1 - Hematoxylin and Eosin (HE) Stained Liver Portal Area Segmentation Using Multi-Scale Receptive Field Convolutional Neural Network
AU - Xiao, Qi En
AU - Chung, Pau Choo
AU - Tsai, Hung Wen
AU - Cheng, Kuo Sheng
AU - Chow, Nan Haw
AU - Juang, Ying Zong
AU - Tsai, Hann Huei
AU - Wang, Cheng Hsiung
AU - Hsieh, Tsan An
N1 - Funding Information:
Manuscript received July 31, 2019; revised October 2, 2019; accepted November 1, 2019. Date of publication November 6, 2019; date of current version December 12, 2019. This work was supported by the Ministry of Science and Technology (MOST), Taiwan, under Grant MOST 107-2634-F-006-004. The computer time and facilities for this work was supported by the National Center for High-Performance Computing, Taiwan. This article was recommended by Guest Editor M. Valle. (Corresponding author: Cheng-Hsiung Wang.) Q.-E. Xiao, P.-C. Chung, and T.-A. Hsieh are with the Institute of Computer and Communication Engineering, National Cheng Kung University, Tainan 70101, Taiwan (e-mail: [email protected]; [email protected]; [email protected]).
Funding Information:
This work was supported by the Ministry of Science and Technology (MOST), Taiwan, under Grant MOST 107-2634-F-006-004. The computer time and facilities for this work was supported by the National Center for High-Performance Computing, Taiwan. This article was recommended by Guest Editor M. Valle.
Publisher Copyright:
© 2011 IEEE.
PY - 2019/12
Y1 - 2019/12
N2 - Portal area segmentation is an important step in the quantitative histological analysis process for hepatitis grading. However, portal areas often appear of different sizes and appearances due to the variations of surrounding components such as the ductule, bile duct, artery, and portal vein. The slim fibrosis expanding from the portal area further increases challenges of the portal area segmentation. A Multi-scale Receptive Field Convolutional Neural Network (MRF-CNN) for the segmentation of the liver portal areas in hematoxylin and eosin (HE) stained whole slide images (WSIs) is proposed in this paper. The MRF-CNN adopts the atrous spatial pyramid pooling (ASPP) with multiple atrous rates and symmetric encoder-decoder with feature concatenation architecture. The atrous rates in ASPP are devised of receptive fields to extract features of meaningful tissue components in parallel in portal areas. Along with the MRF-CNN, a small object sensitive loss function is also proposed to have the network focus on small portal areas and slim fibrosis. The results show that the proposed model achieves Intersection over Union (IOU) of 0.87 and sensitivity of 0.92. Compared to recent segmentation researches such as Fully Convolutional Network (FCN), U-Net and SegNet, the proposed network achieves an overall the best IOU and sensitivity performance. Experimental results also show that the designed ASPP block benefits in feature extraction, and the ability of identifying small objects in proposed small object sensitive loss has a significant improvement of the segmentation result comparing to the original cross entropy loss.
AB - Portal area segmentation is an important step in the quantitative histological analysis process for hepatitis grading. However, portal areas often appear of different sizes and appearances due to the variations of surrounding components such as the ductule, bile duct, artery, and portal vein. The slim fibrosis expanding from the portal area further increases challenges of the portal area segmentation. A Multi-scale Receptive Field Convolutional Neural Network (MRF-CNN) for the segmentation of the liver portal areas in hematoxylin and eosin (HE) stained whole slide images (WSIs) is proposed in this paper. The MRF-CNN adopts the atrous spatial pyramid pooling (ASPP) with multiple atrous rates and symmetric encoder-decoder with feature concatenation architecture. The atrous rates in ASPP are devised of receptive fields to extract features of meaningful tissue components in parallel in portal areas. Along with the MRF-CNN, a small object sensitive loss function is also proposed to have the network focus on small portal areas and slim fibrosis. The results show that the proposed model achieves Intersection over Union (IOU) of 0.87 and sensitivity of 0.92. Compared to recent segmentation researches such as Fully Convolutional Network (FCN), U-Net and SegNet, the proposed network achieves an overall the best IOU and sensitivity performance. Experimental results also show that the designed ASPP block benefits in feature extraction, and the ability of identifying small objects in proposed small object sensitive loss has a significant improvement of the segmentation result comparing to the original cross entropy loss.
UR - http://www.scopus.com/inward/record.url?scp=85074830018&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85074830018&partnerID=8YFLogxK
U2 - 10.1109/JETCAS.2019.2952063
DO - 10.1109/JETCAS.2019.2952063
M3 - Article
AN - SCOPUS:85074830018
SN - 2156-3357
VL - 9
SP - 623
EP - 634
JO - IEEE Journal on Emerging and Selected Topics in Circuits and Systems
JF - IEEE Journal on Emerging and Selected Topics in Circuits and Systems
IS - 4
M1 - 8892562
ER -