Computer-aided assessment of regional abdominal fat with food residue removal in CT

Sokratis Makrogiannis, Giorgio Caturegli, Christos Davatzikos, Luigi Ferrucci

Research output: Contribution to journalArticle

Abstract

Rationale and Objectives: Separate quantification of abdominal subcutaneous and visceral fat regions is essential to understand the role of regional adiposity as risk factor in epidemiological studies. Fat quantification is often based on computed tomography (CT) because fat density is distinct from other tissue densities in the abdomen. However, the presence of intestinal food residues with densities similar to fat may reduce fat quantification accuracy. We introduce an abdominal fat quantification method in CT with interest in food residue removal. Materials and Methods: Total fat was identified in the feature space of Hounsfield units and divided into subcutaneous and visceral components using model-based segmentation. Regions of food residues were identified and removed from visceral fat using a machine learning method integrating intensity, texture, and spatial information. Cost-weighting and bagging techniques were investigated to address class imbalance. Results: We validated our automated food residue removal technique against semimanual quantifications. Our feature selection experiments indicated that joint intensity and texture features produce the highest classification accuracy at 95%. We explored generalization capability using k-fold cross-validation and receiver operating characteristic (ROC) analysis with variable k. Losses in accuracy and area under ROC curve between maximum and minimum k were limited to 0.1% and 0.3%. We validated tissue segmentation against reference semimanual delineations. The Dice similarity scores were as high as 93.1 for subcutaneous fat and 85.6 for visceral fat. Conclusions: Computer-aided regional abdominal fat quantification is a reliable computational tool for large-scale epidemiological studies. Our proposed intestinal food residue reduction scheme is an original contribution of this work. Validation experiments indicate very good accuracy and generalization capability.

Original languageEnglish (US)
Pages (from-to)1413-1421
Number of pages9
JournalAcademic Radiology
Volume20
Issue number11
DOIs
StatePublished - Nov 2013
Externally publishedYes

Fingerprint

Abdominal Fat
Fats
Tomography
Intra-Abdominal Fat
Food
ROC Curve
Epidemiologic Studies
Abdominal Subcutaneous Fat
Subcutaneous Fat
Adiposity
Abdomen
Joints
Costs and Cost Analysis

Keywords

  • Body composition assessment
  • False positive reduction

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Computer-aided assessment of regional abdominal fat with food residue removal in CT. / Makrogiannis, Sokratis; Caturegli, Giorgio; Davatzikos, Christos; Ferrucci, Luigi.

In: Academic Radiology, Vol. 20, No. 11, 11.2013, p. 1413-1421.

Research output: Contribution to journalArticle

Makrogiannis, S, Caturegli, G, Davatzikos, C & Ferrucci, L 2013, 'Computer-aided assessment of regional abdominal fat with food residue removal in CT', Academic Radiology, vol. 20, no. 11, pp. 1413-1421. https://doi.org/10.1016/j.acra.2013.08.007
Makrogiannis, Sokratis ; Caturegli, Giorgio ; Davatzikos, Christos ; Ferrucci, Luigi. / Computer-aided assessment of regional abdominal fat with food residue removal in CT. In: Academic Radiology. 2013 ; Vol. 20, No. 11. pp. 1413-1421.
@article{20ba7c989a784c13a010e5313f74d05f,
title = "Computer-aided assessment of regional abdominal fat with food residue removal in CT",
abstract = "Rationale and Objectives: Separate quantification of abdominal subcutaneous and visceral fat regions is essential to understand the role of regional adiposity as risk factor in epidemiological studies. Fat quantification is often based on computed tomography (CT) because fat density is distinct from other tissue densities in the abdomen. However, the presence of intestinal food residues with densities similar to fat may reduce fat quantification accuracy. We introduce an abdominal fat quantification method in CT with interest in food residue removal. Materials and Methods: Total fat was identified in the feature space of Hounsfield units and divided into subcutaneous and visceral components using model-based segmentation. Regions of food residues were identified and removed from visceral fat using a machine learning method integrating intensity, texture, and spatial information. Cost-weighting and bagging techniques were investigated to address class imbalance. Results: We validated our automated food residue removal technique against semimanual quantifications. Our feature selection experiments indicated that joint intensity and texture features produce the highest classification accuracy at 95{\%}. We explored generalization capability using k-fold cross-validation and receiver operating characteristic (ROC) analysis with variable k. Losses in accuracy and area under ROC curve between maximum and minimum k were limited to 0.1{\%} and 0.3{\%}. We validated tissue segmentation against reference semimanual delineations. The Dice similarity scores were as high as 93.1 for subcutaneous fat and 85.6 for visceral fat. Conclusions: Computer-aided regional abdominal fat quantification is a reliable computational tool for large-scale epidemiological studies. Our proposed intestinal food residue reduction scheme is an original contribution of this work. Validation experiments indicate very good accuracy and generalization capability.",
keywords = "Body composition assessment, False positive reduction",
author = "Sokratis Makrogiannis and Giorgio Caturegli and Christos Davatzikos and Luigi Ferrucci",
year = "2013",
month = "11",
doi = "10.1016/j.acra.2013.08.007",
language = "English (US)",
volume = "20",
pages = "1413--1421",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "11",

}

TY - JOUR

T1 - Computer-aided assessment of regional abdominal fat with food residue removal in CT

AU - Makrogiannis, Sokratis

AU - Caturegli, Giorgio

AU - Davatzikos, Christos

AU - Ferrucci, Luigi

PY - 2013/11

Y1 - 2013/11

N2 - Rationale and Objectives: Separate quantification of abdominal subcutaneous and visceral fat regions is essential to understand the role of regional adiposity as risk factor in epidemiological studies. Fat quantification is often based on computed tomography (CT) because fat density is distinct from other tissue densities in the abdomen. However, the presence of intestinal food residues with densities similar to fat may reduce fat quantification accuracy. We introduce an abdominal fat quantification method in CT with interest in food residue removal. Materials and Methods: Total fat was identified in the feature space of Hounsfield units and divided into subcutaneous and visceral components using model-based segmentation. Regions of food residues were identified and removed from visceral fat using a machine learning method integrating intensity, texture, and spatial information. Cost-weighting and bagging techniques were investigated to address class imbalance. Results: We validated our automated food residue removal technique against semimanual quantifications. Our feature selection experiments indicated that joint intensity and texture features produce the highest classification accuracy at 95%. We explored generalization capability using k-fold cross-validation and receiver operating characteristic (ROC) analysis with variable k. Losses in accuracy and area under ROC curve between maximum and minimum k were limited to 0.1% and 0.3%. We validated tissue segmentation against reference semimanual delineations. The Dice similarity scores were as high as 93.1 for subcutaneous fat and 85.6 for visceral fat. Conclusions: Computer-aided regional abdominal fat quantification is a reliable computational tool for large-scale epidemiological studies. Our proposed intestinal food residue reduction scheme is an original contribution of this work. Validation experiments indicate very good accuracy and generalization capability.

AB - Rationale and Objectives: Separate quantification of abdominal subcutaneous and visceral fat regions is essential to understand the role of regional adiposity as risk factor in epidemiological studies. Fat quantification is often based on computed tomography (CT) because fat density is distinct from other tissue densities in the abdomen. However, the presence of intestinal food residues with densities similar to fat may reduce fat quantification accuracy. We introduce an abdominal fat quantification method in CT with interest in food residue removal. Materials and Methods: Total fat was identified in the feature space of Hounsfield units and divided into subcutaneous and visceral components using model-based segmentation. Regions of food residues were identified and removed from visceral fat using a machine learning method integrating intensity, texture, and spatial information. Cost-weighting and bagging techniques were investigated to address class imbalance. Results: We validated our automated food residue removal technique against semimanual quantifications. Our feature selection experiments indicated that joint intensity and texture features produce the highest classification accuracy at 95%. We explored generalization capability using k-fold cross-validation and receiver operating characteristic (ROC) analysis with variable k. Losses in accuracy and area under ROC curve between maximum and minimum k were limited to 0.1% and 0.3%. We validated tissue segmentation against reference semimanual delineations. The Dice similarity scores were as high as 93.1 for subcutaneous fat and 85.6 for visceral fat. Conclusions: Computer-aided regional abdominal fat quantification is a reliable computational tool for large-scale epidemiological studies. Our proposed intestinal food residue reduction scheme is an original contribution of this work. Validation experiments indicate very good accuracy and generalization capability.

KW - Body composition assessment

KW - False positive reduction

UR - http://www.scopus.com/inward/record.url?scp=84885357878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885357878&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2013.08.007

DO - 10.1016/j.acra.2013.08.007

M3 - Article

C2 - 24119354

AN - SCOPUS:84885357878

VL - 20

SP - 1413

EP - 1421

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 11

ER -