Automated abdominal fat quantification and food residue removal in CT

Sokratis Makrogiannis, Ramona Ramachandran, Chee W. Chia, Luigi Ferrucci

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Quantification of distinct subcutaneous and visceral fat regions in the abdomen is essential in clinical studies of metabolic disorders and cardiovascular disease. Computed Tomography (CT) is a widely adopted imaging technology for abdominal fat quantification because the intensity range of fat in Hounsfield Units (HU) is distinct from other tissues in the pelvis and abdomen. Nevertheless, it has been observed that the quantification of visceral fat based solely on intensity is subject to errors caused by food residues in the intestines that may have intensities similar to fat. Herein we present a method for automated quantification of abdominal fat in CT with emphasis on reducing errors in visceral fat measurements caused by food residues. The fat pixels are first identified in the feature space of HUs and then divided into subcutaneous and visceral component using anatomic location. Food residues within the intestines that are previously inaccurately labeled as visceral fat (false positives) are identified and removed using a machine learning technique. Experimental results include validation against reference data over 144 CT images to test the generalization capability of our scheme.

Original languageEnglish (US)
Title of host publicationProceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis
Pages81-86
Number of pages6
DOIs
StatePublished - 2012
Externally publishedYes
Event2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis, MMBIA 2012 - Breckenridge, CO, United States
Duration: Jan 9 2012Jan 10 2012

Other

Other2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis, MMBIA 2012
CountryUnited States
CityBreckenridge, CO
Period1/9/121/10/12

Fingerprint

Abdominal Fat
Intra-Abdominal Fat
Computed Tomography
Oils and fats
Quantification
Tomography
Food
Fats
Abdomen
Intestines
Distinct
Subcutaneous Fat
Metabolic Diseases
Feature Space
Pelvis
False Positive
Disorder
Machine Learning
Cardiovascular Diseases
Pixel

ASJC Scopus subject areas

  • Applied Mathematics
  • Radiology Nuclear Medicine and imaging
  • Biomedical Engineering

Cite this

Makrogiannis, S., Ramachandran, R., Chia, C. W., & Ferrucci, L. (2012). Automated abdominal fat quantification and food residue removal in CT. In Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis (pp. 81-86). [6164738] https://doi.org/10.1109/MMBIA.2012.6164738

Automated abdominal fat quantification and food residue removal in CT. / Makrogiannis, Sokratis; Ramachandran, Ramona; Chia, Chee W.; Ferrucci, Luigi.

Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis. 2012. p. 81-86 6164738.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Makrogiannis, S, Ramachandran, R, Chia, CW & Ferrucci, L 2012, Automated abdominal fat quantification and food residue removal in CT. in Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis., 6164738, pp. 81-86, 2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis, MMBIA 2012, Breckenridge, CO, United States, 1/9/12. https://doi.org/10.1109/MMBIA.2012.6164738
Makrogiannis S, Ramachandran R, Chia CW, Ferrucci L. Automated abdominal fat quantification and food residue removal in CT. In Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis. 2012. p. 81-86. 6164738 https://doi.org/10.1109/MMBIA.2012.6164738
Makrogiannis, Sokratis ; Ramachandran, Ramona ; Chia, Chee W. ; Ferrucci, Luigi. / Automated abdominal fat quantification and food residue removal in CT. Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis. 2012. pp. 81-86
@inproceedings{bf39f0db4bda4123a74876a0311813c7,
title = "Automated abdominal fat quantification and food residue removal in CT",
abstract = "Quantification of distinct subcutaneous and visceral fat regions in the abdomen is essential in clinical studies of metabolic disorders and cardiovascular disease. Computed Tomography (CT) is a widely adopted imaging technology for abdominal fat quantification because the intensity range of fat in Hounsfield Units (HU) is distinct from other tissues in the pelvis and abdomen. Nevertheless, it has been observed that the quantification of visceral fat based solely on intensity is subject to errors caused by food residues in the intestines that may have intensities similar to fat. Herein we present a method for automated quantification of abdominal fat in CT with emphasis on reducing errors in visceral fat measurements caused by food residues. The fat pixels are first identified in the feature space of HUs and then divided into subcutaneous and visceral component using anatomic location. Food residues within the intestines that are previously inaccurately labeled as visceral fat (false positives) are identified and removed using a machine learning technique. Experimental results include validation against reference data over 144 CT images to test the generalization capability of our scheme.",
author = "Sokratis Makrogiannis and Ramona Ramachandran and Chia, {Chee W.} and Luigi Ferrucci",
year = "2012",
doi = "10.1109/MMBIA.2012.6164738",
language = "English (US)",
isbn = "9781467303521",
pages = "81--86",
booktitle = "Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis",

}

TY - GEN

T1 - Automated abdominal fat quantification and food residue removal in CT

AU - Makrogiannis, Sokratis

AU - Ramachandran, Ramona

AU - Chia, Chee W.

AU - Ferrucci, Luigi

PY - 2012

Y1 - 2012

N2 - Quantification of distinct subcutaneous and visceral fat regions in the abdomen is essential in clinical studies of metabolic disorders and cardiovascular disease. Computed Tomography (CT) is a widely adopted imaging technology for abdominal fat quantification because the intensity range of fat in Hounsfield Units (HU) is distinct from other tissues in the pelvis and abdomen. Nevertheless, it has been observed that the quantification of visceral fat based solely on intensity is subject to errors caused by food residues in the intestines that may have intensities similar to fat. Herein we present a method for automated quantification of abdominal fat in CT with emphasis on reducing errors in visceral fat measurements caused by food residues. The fat pixels are first identified in the feature space of HUs and then divided into subcutaneous and visceral component using anatomic location. Food residues within the intestines that are previously inaccurately labeled as visceral fat (false positives) are identified and removed using a machine learning technique. Experimental results include validation against reference data over 144 CT images to test the generalization capability of our scheme.

AB - Quantification of distinct subcutaneous and visceral fat regions in the abdomen is essential in clinical studies of metabolic disorders and cardiovascular disease. Computed Tomography (CT) is a widely adopted imaging technology for abdominal fat quantification because the intensity range of fat in Hounsfield Units (HU) is distinct from other tissues in the pelvis and abdomen. Nevertheless, it has been observed that the quantification of visceral fat based solely on intensity is subject to errors caused by food residues in the intestines that may have intensities similar to fat. Herein we present a method for automated quantification of abdominal fat in CT with emphasis on reducing errors in visceral fat measurements caused by food residues. The fat pixels are first identified in the feature space of HUs and then divided into subcutaneous and visceral component using anatomic location. Food residues within the intestines that are previously inaccurately labeled as visceral fat (false positives) are identified and removed using a machine learning technique. Experimental results include validation against reference data over 144 CT images to test the generalization capability of our scheme.

UR - http://www.scopus.com/inward/record.url?scp=84859914214&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859914214&partnerID=8YFLogxK

U2 - 10.1109/MMBIA.2012.6164738

DO - 10.1109/MMBIA.2012.6164738

M3 - Conference contribution

AN - SCOPUS:84859914214

SN - 9781467303521

SP - 81

EP - 86

BT - Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis

ER -