Automated assessment of steatosis in murine fatty liver

Deepak Sethunath, Siripriya Morusu, Mihran Tuceryan, Oscar Cummings, Hao Zhang, Xiao-Ming Yin, Scott Vanderbeck, Naga Chalasani, Samer Gawrieh

Research output: Contribution to journalArticle

Abstract

Although mice are commonly used to study different aspects of fatty liver disease, currently there are no validated fully automated methods to assess steatosis in mice. Accurate detection of macro- and microsteatosis in murine models of fatty liver disease is important in studying disease pathogenesis and detecting potential hepatotoxic signature during drug development. Further, precise quantification of macrosteatosis is essential for quantifying effects of therapies. Here, we develop and validate the performance of automated classifiers built using image processing and machine learning methods for detection of macro- and microsteatosis in murine fatty liver disease and study the correlation of automated quantification of macrosteatosis with expert pathologist’s semi-quantitative grades. The analysis is performed on digital images of 27 Hematoxylin & Eosin stained murine liver biopsy samples. An expert liver pathologist scored the amount of macrosteatosis and also annotated macro- and microsteatosis lesions on the biopsy images using a web-application. Using these annotations, supervised machine learning and image processing techniques, we created classifiers to detect macro- and microsteatosis. For macrosteatosis prediction, the model’s precision, sensitivity and area under the receiver operator characteristic (AUROC) were 94.2%, 95%, 99.1% respectively. When correlated with pathologist’s semi-quantitative grade of steatosis, the model fits with a coefficient of determination value of 0.905. For microsteatosis prediction, the model has precision, sensitivity and AUROC of 79.2%, 77%, 78.1% respectively. Validation by the expert pathologist of classifier’s predictions made on unseen images of biopsy samples showed 100% and 63% accuracy for macro- and microsteatosis, respectively. This novel work demonstrates that fully automated assessment of steatosis is feasible in murine liver biopsies images. Our classifier has excellent sensitivity and accuracy for detection of macrosteatosis in murine fatty liver disease.

Original languageEnglish (US)
Article numbere0197242
JournalPLoS One
Volume13
Issue number5
DOIs
StatePublished - May 1 2018

Fingerprint

fatty liver
Fatty Liver
Liver
Liver Diseases
liver diseases
Biopsy
Macros
biopsy
mice
Classifiers
artificial intelligence
liver
prediction
Hematoxylin
Eosine Yellowish-(YS)
Learning systems
image analysis
Image processing
digital images
lesions (animal)

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Automated assessment of steatosis in murine fatty liver. / Sethunath, Deepak; Morusu, Siripriya; Tuceryan, Mihran; Cummings, Oscar; Zhang, Hao; Yin, Xiao-Ming; Vanderbeck, Scott; Chalasani, Naga; Gawrieh, Samer.

In: PLoS One, Vol. 13, No. 5, e0197242, 01.05.2018.

Research output: Contribution to journalArticle

Sethunath, Deepak ; Morusu, Siripriya ; Tuceryan, Mihran ; Cummings, Oscar ; Zhang, Hao ; Yin, Xiao-Ming ; Vanderbeck, Scott ; Chalasani, Naga ; Gawrieh, Samer. / Automated assessment of steatosis in murine fatty liver. In: PLoS One. 2018 ; Vol. 13, No. 5.
@article{dbdc5973a3024ac2b8b281a67f948a11,
title = "Automated assessment of steatosis in murine fatty liver",
abstract = "Although mice are commonly used to study different aspects of fatty liver disease, currently there are no validated fully automated methods to assess steatosis in mice. Accurate detection of macro- and microsteatosis in murine models of fatty liver disease is important in studying disease pathogenesis and detecting potential hepatotoxic signature during drug development. Further, precise quantification of macrosteatosis is essential for quantifying effects of therapies. Here, we develop and validate the performance of automated classifiers built using image processing and machine learning methods for detection of macro- and microsteatosis in murine fatty liver disease and study the correlation of automated quantification of macrosteatosis with expert pathologist’s semi-quantitative grades. The analysis is performed on digital images of 27 Hematoxylin & Eosin stained murine liver biopsy samples. An expert liver pathologist scored the amount of macrosteatosis and also annotated macro- and microsteatosis lesions on the biopsy images using a web-application. Using these annotations, supervised machine learning and image processing techniques, we created classifiers to detect macro- and microsteatosis. For macrosteatosis prediction, the model’s precision, sensitivity and area under the receiver operator characteristic (AUROC) were 94.2{\%}, 95{\%}, 99.1{\%} respectively. When correlated with pathologist’s semi-quantitative grade of steatosis, the model fits with a coefficient of determination value of 0.905. For microsteatosis prediction, the model has precision, sensitivity and AUROC of 79.2{\%}, 77{\%}, 78.1{\%} respectively. Validation by the expert pathologist of classifier’s predictions made on unseen images of biopsy samples showed 100{\%} and 63{\%} accuracy for macro- and microsteatosis, respectively. This novel work demonstrates that fully automated assessment of steatosis is feasible in murine liver biopsies images. Our classifier has excellent sensitivity and accuracy for detection of macrosteatosis in murine fatty liver disease.",
author = "Deepak Sethunath and Siripriya Morusu and Mihran Tuceryan and Oscar Cummings and Hao Zhang and Xiao-Ming Yin and Scott Vanderbeck and Naga Chalasani and Samer Gawrieh",
year = "2018",
month = "5",
day = "1",
doi = "10.1371/journal.pone.0197242",
language = "English (US)",
volume = "13",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "5",

}

TY - JOUR

T1 - Automated assessment of steatosis in murine fatty liver

AU - Sethunath, Deepak

AU - Morusu, Siripriya

AU - Tuceryan, Mihran

AU - Cummings, Oscar

AU - Zhang, Hao

AU - Yin, Xiao-Ming

AU - Vanderbeck, Scott

AU - Chalasani, Naga

AU - Gawrieh, Samer

PY - 2018/5/1

Y1 - 2018/5/1

N2 - Although mice are commonly used to study different aspects of fatty liver disease, currently there are no validated fully automated methods to assess steatosis in mice. Accurate detection of macro- and microsteatosis in murine models of fatty liver disease is important in studying disease pathogenesis and detecting potential hepatotoxic signature during drug development. Further, precise quantification of macrosteatosis is essential for quantifying effects of therapies. Here, we develop and validate the performance of automated classifiers built using image processing and machine learning methods for detection of macro- and microsteatosis in murine fatty liver disease and study the correlation of automated quantification of macrosteatosis with expert pathologist’s semi-quantitative grades. The analysis is performed on digital images of 27 Hematoxylin & Eosin stained murine liver biopsy samples. An expert liver pathologist scored the amount of macrosteatosis and also annotated macro- and microsteatosis lesions on the biopsy images using a web-application. Using these annotations, supervised machine learning and image processing techniques, we created classifiers to detect macro- and microsteatosis. For macrosteatosis prediction, the model’s precision, sensitivity and area under the receiver operator characteristic (AUROC) were 94.2%, 95%, 99.1% respectively. When correlated with pathologist’s semi-quantitative grade of steatosis, the model fits with a coefficient of determination value of 0.905. For microsteatosis prediction, the model has precision, sensitivity and AUROC of 79.2%, 77%, 78.1% respectively. Validation by the expert pathologist of classifier’s predictions made on unseen images of biopsy samples showed 100% and 63% accuracy for macro- and microsteatosis, respectively. This novel work demonstrates that fully automated assessment of steatosis is feasible in murine liver biopsies images. Our classifier has excellent sensitivity and accuracy for detection of macrosteatosis in murine fatty liver disease.

AB - Although mice are commonly used to study different aspects of fatty liver disease, currently there are no validated fully automated methods to assess steatosis in mice. Accurate detection of macro- and microsteatosis in murine models of fatty liver disease is important in studying disease pathogenesis and detecting potential hepatotoxic signature during drug development. Further, precise quantification of macrosteatosis is essential for quantifying effects of therapies. Here, we develop and validate the performance of automated classifiers built using image processing and machine learning methods for detection of macro- and microsteatosis in murine fatty liver disease and study the correlation of automated quantification of macrosteatosis with expert pathologist’s semi-quantitative grades. The analysis is performed on digital images of 27 Hematoxylin & Eosin stained murine liver biopsy samples. An expert liver pathologist scored the amount of macrosteatosis and also annotated macro- and microsteatosis lesions on the biopsy images using a web-application. Using these annotations, supervised machine learning and image processing techniques, we created classifiers to detect macro- and microsteatosis. For macrosteatosis prediction, the model’s precision, sensitivity and area under the receiver operator characteristic (AUROC) were 94.2%, 95%, 99.1% respectively. When correlated with pathologist’s semi-quantitative grade of steatosis, the model fits with a coefficient of determination value of 0.905. For microsteatosis prediction, the model has precision, sensitivity and AUROC of 79.2%, 77%, 78.1% respectively. Validation by the expert pathologist of classifier’s predictions made on unseen images of biopsy samples showed 100% and 63% accuracy for macro- and microsteatosis, respectively. This novel work demonstrates that fully automated assessment of steatosis is feasible in murine liver biopsies images. Our classifier has excellent sensitivity and accuracy for detection of macrosteatosis in murine fatty liver disease.

UR - http://www.scopus.com/inward/record.url?scp=85046836441&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046836441&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0197242

DO - 10.1371/journal.pone.0197242

M3 - Article

VL - 13

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 5

M1 - e0197242

ER -