Liver steatosis segmentation with deep learning methods

Xiaoyuan Guo; Fusheng Wang; George Teodoro; Alton B. Farris; Jun Kong

doi:10.1109/ISBI.2019.8759600

Liver steatosis segmentation with deep learning methods

Xiaoyuan Guo
, Fusheng Wang
, George Teodoro
, Alton B. Farris
, Jun Kong

Emory University
Universidade de Brasília
Georgia State University
Stony Brook University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

24 Scopus citations

Abstract

Liver steatosis is known as the abnormal accumulation of lipids within cells. An accurate quantification of steatosis area within the liver histopathological microscopy images plays an important role in liver disease diagnosis and transplantation assessment. Such a quantification analysis often requires a precise steatosis segmentation that is challenging due to abundant presence of highly overlapped steatosis droplets. In this paper, a deep learning model Mask-RCNN is used to segment the steatosis droplets in clumps. Extended from Faster R-CNN, Mask-RCNN can predict object masks in addition to bounding box detection. With transfer learning, the resulting model is able to segment overlapped steatosis regions at 75.87% by Average Precision, 60.66% by Recall, 65.88% by F1-score, and 76.97% by Jaccard index, promising to support liver disease diagnosis and allograft rejection prediction in future clinical practice.

Original language	English
Title of host publication	ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
Publisher	IEEE Computer Society
Pages	24-27
Number of pages	4
ISBN (Electronic)	9781538636411
DOIs	https://doi.org/10.1109/ISBI.2019.8759600
State	Published - Apr 2019
Event	16th IEEE International Symposium on Biomedical Imaging, ISBI 2019 - Venice, Italy Duration: Apr 8 2019 → Apr 11 2019

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2019-April
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
Country/Territory	Italy
City	Venice
Period	04/8/19 → 04/11/19

Keywords

Deep learning
Liver steatosis segmentation
Whole-slide microscopy image

Access to Document

10.1109/ISBI.2019.8759600

Cite this

@inproceedings{f4cc8813b27145ef836f8efd6cbc7154,

title = "Liver steatosis segmentation with deep learning methods",

abstract = "Liver steatosis is known as the abnormal accumulation of lipids within cells. An accurate quantification of steatosis area within the liver histopathological microscopy images plays an important role in liver disease diagnosis and transplantation assessment. Such a quantification analysis often requires a precise steatosis segmentation that is challenging due to abundant presence of highly overlapped steatosis droplets. In this paper, a deep learning model Mask-RCNN is used to segment the steatosis droplets in clumps. Extended from Faster R-CNN, Mask-RCNN can predict object masks in addition to bounding box detection. With transfer learning, the resulting model is able to segment overlapped steatosis regions at 75.87\% by Average Precision, 60.66\% by Recall, 65.88\% by F1-score, and 76.97\% by Jaccard index, promising to support liver disease diagnosis and allograft rejection prediction in future clinical practice.",

keywords = "Deep learning, Liver steatosis segmentation, Whole-slide microscopy image",

author = "Xiaoyuan Guo and Fusheng Wang and George Teodoro and Farris, \{Alton B.\} and Jun Kong",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019 ; Conference date: 08-04-2019 Through 11-04-2019",

year = "2019",

month = apr,

doi = "10.1109/ISBI.2019.8759600",

language = "English",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

pages = "24--27",

booktitle = "ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging",

}

Guo, X, Wang, F, Teodoro, G, Farris, AB & Kong, J 2019, Liver steatosis segmentation with deep learning methods. in ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging., 8759600, Proceedings - International Symposium on Biomedical Imaging, vol. 2019-April, IEEE Computer Society, pp. 24-27, 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019, Venice, Italy, 04/8/19. https://doi.org/10.1109/ISBI.2019.8759600

Liver steatosis segmentation with deep learning methods. / Guo, Xiaoyuan; Wang, Fusheng; Teodoro, George et al.
ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2019. p. 24-27 8759600 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2019-April).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Liver steatosis segmentation with deep learning methods

AU - Guo, Xiaoyuan

AU - Wang, Fusheng

AU - Teodoro, George

AU - Farris, Alton B.

AU - Kong, Jun

PY - 2019/4

Y1 - 2019/4

N2 - Liver steatosis is known as the abnormal accumulation of lipids within cells. An accurate quantification of steatosis area within the liver histopathological microscopy images plays an important role in liver disease diagnosis and transplantation assessment. Such a quantification analysis often requires a precise steatosis segmentation that is challenging due to abundant presence of highly overlapped steatosis droplets. In this paper, a deep learning model Mask-RCNN is used to segment the steatosis droplets in clumps. Extended from Faster R-CNN, Mask-RCNN can predict object masks in addition to bounding box detection. With transfer learning, the resulting model is able to segment overlapped steatosis regions at 75.87% by Average Precision, 60.66% by Recall, 65.88% by F1-score, and 76.97% by Jaccard index, promising to support liver disease diagnosis and allograft rejection prediction in future clinical practice.

AB - Liver steatosis is known as the abnormal accumulation of lipids within cells. An accurate quantification of steatosis area within the liver histopathological microscopy images plays an important role in liver disease diagnosis and transplantation assessment. Such a quantification analysis often requires a precise steatosis segmentation that is challenging due to abundant presence of highly overlapped steatosis droplets. In this paper, a deep learning model Mask-RCNN is used to segment the steatosis droplets in clumps. Extended from Faster R-CNN, Mask-RCNN can predict object masks in addition to bounding box detection. With transfer learning, the resulting model is able to segment overlapped steatosis regions at 75.87% by Average Precision, 60.66% by Recall, 65.88% by F1-score, and 76.97% by Jaccard index, promising to support liver disease diagnosis and allograft rejection prediction in future clinical practice.

KW - Deep learning

KW - Liver steatosis segmentation

KW - Whole-slide microscopy image

UR - https://www.scopus.com/pages/publications/85073890255

U2 - 10.1109/ISBI.2019.8759600

DO - 10.1109/ISBI.2019.8759600

M3 - Conference contribution

AN - SCOPUS:85073890255

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 24

EP - 27

BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

T2 - 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019

Y2 - 8 April 2019 through 11 April 2019

ER -

Liver steatosis segmentation with deep learning methods

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this