Decision support system for fatty liver disease using GIST descriptors extracted from ultrasound images

U. Rajendra Acharya, Hamido Fujita, Shreya Bhat, U. Raghavendra, Anjan Gudigar, Filippo Molinari, Anushya Vijayananthan, Kwan Hoong Ng

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Steatosis or fatty liver disease (FLD) is characterized by the abnormal retention of large vacuoles of neutral fat in the liver cells, either due to alcoholism or metabolic syndrome. Succession of FLD can lead to severe liver diseases such as hepatocellular carcinoma, cirrhosis and hepatic inflammation but it is a reversible disease if diagnosed early. Thus, computer-aided diagnostic tools play a very important role in the automated diagnosis of FLD. This paper focuses on the detection of steatosis and classification of steatotic livers from the normal using ultrasound images. The significant information from the image is extracted using GIST descriptor models. Marginal Fisher Analysis (MFA) integrated with Wilcoxon signed-rank test helps to eliminate the trivial features and provides the distinctive features for qualitative classification. Finally the clinically significant features are fused using classifiers such as decision tree (DT), support vector machine (SVM), adaBoost, k-nearest neighbor (kNN), probabilistic neural network (PNN), naïve Bayes (NB), fuzzy Sugeno (FS), linear and quadratic discriminant analysis classification of normal and abnormal liver images. Results portray that PNN classifier can diagnose FLD with an average classification accuracy of 98%, 96% sensitivity, 100% specificity and Area Under Curve (AUC) of 0.9674 correctly.

Original languageEnglish
Pages (from-to)32-39
Number of pages8
JournalInformation Fusion
Volume29
DOIs
Publication statusPublished - 01-05-2016

Fingerprint

Decision support systems
Liver
Ultrasonics
Classifiers
Neural networks
Adaptive boosting
Discriminant analysis
Decision trees
Oils and fats
Support vector machines

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Information Systems
  • Hardware and Architecture

Cite this

Acharya, U. Rajendra ; Fujita, Hamido ; Bhat, Shreya ; Raghavendra, U. ; Gudigar, Anjan ; Molinari, Filippo ; Vijayananthan, Anushya ; Hoong Ng, Kwan. / Decision support system for fatty liver disease using GIST descriptors extracted from ultrasound images. In: Information Fusion. 2016 ; Vol. 29. pp. 32-39.
@article{d17b897cf6e2438fba48134d42148304,
title = "Decision support system for fatty liver disease using GIST descriptors extracted from ultrasound images",
abstract = "Steatosis or fatty liver disease (FLD) is characterized by the abnormal retention of large vacuoles of neutral fat in the liver cells, either due to alcoholism or metabolic syndrome. Succession of FLD can lead to severe liver diseases such as hepatocellular carcinoma, cirrhosis and hepatic inflammation but it is a reversible disease if diagnosed early. Thus, computer-aided diagnostic tools play a very important role in the automated diagnosis of FLD. This paper focuses on the detection of steatosis and classification of steatotic livers from the normal using ultrasound images. The significant information from the image is extracted using GIST descriptor models. Marginal Fisher Analysis (MFA) integrated with Wilcoxon signed-rank test helps to eliminate the trivial features and provides the distinctive features for qualitative classification. Finally the clinically significant features are fused using classifiers such as decision tree (DT), support vector machine (SVM), adaBoost, k-nearest neighbor (kNN), probabilistic neural network (PNN), na{\"i}ve Bayes (NB), fuzzy Sugeno (FS), linear and quadratic discriminant analysis classification of normal and abnormal liver images. Results portray that PNN classifier can diagnose FLD with an average classification accuracy of 98{\%}, 96{\%} sensitivity, 100{\%} specificity and Area Under Curve (AUC) of 0.9674 correctly.",
author = "Acharya, {U. Rajendra} and Hamido Fujita and Shreya Bhat and U. Raghavendra and Anjan Gudigar and Filippo Molinari and Anushya Vijayananthan and {Hoong Ng}, Kwan",
year = "2016",
month = "5",
day = "1",
doi = "10.1016/j.inffus.2015.09.006",
language = "English",
volume = "29",
pages = "32--39",
journal = "Information Fusion",
issn = "1566-2535",
publisher = "Elsevier",

}

Decision support system for fatty liver disease using GIST descriptors extracted from ultrasound images. / Acharya, U. Rajendra; Fujita, Hamido; Bhat, Shreya; Raghavendra, U.; Gudigar, Anjan; Molinari, Filippo; Vijayananthan, Anushya; Hoong Ng, Kwan.

In: Information Fusion, Vol. 29, 01.05.2016, p. 32-39.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Decision support system for fatty liver disease using GIST descriptors extracted from ultrasound images

AU - Acharya, U. Rajendra

AU - Fujita, Hamido

AU - Bhat, Shreya

AU - Raghavendra, U.

AU - Gudigar, Anjan

AU - Molinari, Filippo

AU - Vijayananthan, Anushya

AU - Hoong Ng, Kwan

PY - 2016/5/1

Y1 - 2016/5/1

N2 - Steatosis or fatty liver disease (FLD) is characterized by the abnormal retention of large vacuoles of neutral fat in the liver cells, either due to alcoholism or metabolic syndrome. Succession of FLD can lead to severe liver diseases such as hepatocellular carcinoma, cirrhosis and hepatic inflammation but it is a reversible disease if diagnosed early. Thus, computer-aided diagnostic tools play a very important role in the automated diagnosis of FLD. This paper focuses on the detection of steatosis and classification of steatotic livers from the normal using ultrasound images. The significant information from the image is extracted using GIST descriptor models. Marginal Fisher Analysis (MFA) integrated with Wilcoxon signed-rank test helps to eliminate the trivial features and provides the distinctive features for qualitative classification. Finally the clinically significant features are fused using classifiers such as decision tree (DT), support vector machine (SVM), adaBoost, k-nearest neighbor (kNN), probabilistic neural network (PNN), naïve Bayes (NB), fuzzy Sugeno (FS), linear and quadratic discriminant analysis classification of normal and abnormal liver images. Results portray that PNN classifier can diagnose FLD with an average classification accuracy of 98%, 96% sensitivity, 100% specificity and Area Under Curve (AUC) of 0.9674 correctly.

AB - Steatosis or fatty liver disease (FLD) is characterized by the abnormal retention of large vacuoles of neutral fat in the liver cells, either due to alcoholism or metabolic syndrome. Succession of FLD can lead to severe liver diseases such as hepatocellular carcinoma, cirrhosis and hepatic inflammation but it is a reversible disease if diagnosed early. Thus, computer-aided diagnostic tools play a very important role in the automated diagnosis of FLD. This paper focuses on the detection of steatosis and classification of steatotic livers from the normal using ultrasound images. The significant information from the image is extracted using GIST descriptor models. Marginal Fisher Analysis (MFA) integrated with Wilcoxon signed-rank test helps to eliminate the trivial features and provides the distinctive features for qualitative classification. Finally the clinically significant features are fused using classifiers such as decision tree (DT), support vector machine (SVM), adaBoost, k-nearest neighbor (kNN), probabilistic neural network (PNN), naïve Bayes (NB), fuzzy Sugeno (FS), linear and quadratic discriminant analysis classification of normal and abnormal liver images. Results portray that PNN classifier can diagnose FLD with an average classification accuracy of 98%, 96% sensitivity, 100% specificity and Area Under Curve (AUC) of 0.9674 correctly.

UR - http://www.scopus.com/inward/record.url?scp=84944790601&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944790601&partnerID=8YFLogxK

U2 - 10.1016/j.inffus.2015.09.006

DO - 10.1016/j.inffus.2015.09.006

M3 - Article

VL - 29

SP - 32

EP - 39

JO - Information Fusion

JF - Information Fusion

SN - 1566-2535

ER -