Image-based tissue distribution modeling for skeletal muscle quality characterization

S. Makrogiannis, K. W. Fishbein, A. Z. Moore, R. G. Spencer, L. Ferrucci

Research output: Contribution to journalArticle

Abstract

The identification and characterization of regional body tissues is essential to understand changes that occur with aging and age-related metabolic diseases such as diabetes and obesity and how these diseases affect trajectories of health and functional status. Imaging technologies are frequently used to derive volumetric, area, and density measurements of different tissues. Despite the significance and direct applicability of automated tissue quantification and characterization techniques, these topics have remained relatively underexplored in the medical image analysis literature. We present a method for identification and characterization of muscle and adipose tissue in the midthigh region using MRI. We propose an image-based muscle quality prediction technique that estimates tissue-specific probability density models and their eigenstructures in the joint domain of water- and fat-suppressed voxel signal intensities along with volumetric and intensity-based tissue characteristics computed during the quantification stage. We evaluated the predictive capability of our approach against reference biomechanical muscle quality (MQ) measurements using statistical tests and classification performance experiments. The reference standard for MQ is defined as the ratio of muscle strength to muscle mass. The results show promise for the development of noninvasive image-based MQ descriptors.

Original languageEnglish (US)
Article number7229301
Pages (from-to)805-813
Number of pages9
JournalIEEE Transactions on Biomedical Engineering
Volume63
Issue number4
DOIs
StatePublished - Apr 1 2016
Externally publishedYes

Fingerprint

Muscle
Tissue
Statistical tests
Medical problems
Oils and fats
Magnetic resonance imaging
Image analysis
Aging of materials
Trajectories
Health
Imaging techniques
Water
Experiments

Keywords

  • Magnetic resonance imaging (MRI)
  • Probabilistic modeling
  • Tissue identification and characterization

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Makrogiannis, S., Fishbein, K. W., Moore, A. Z., Spencer, R. G., & Ferrucci, L. (2016). Image-based tissue distribution modeling for skeletal muscle quality characterization. IEEE Transactions on Biomedical Engineering, 63(4), 805-813. [7229301]. https://doi.org/10.1109/TBME.2015.2474305

Image-based tissue distribution modeling for skeletal muscle quality characterization. / Makrogiannis, S.; Fishbein, K. W.; Moore, A. Z.; Spencer, R. G.; Ferrucci, L.

In: IEEE Transactions on Biomedical Engineering, Vol. 63, No. 4, 7229301, 01.04.2016, p. 805-813.

Research output: Contribution to journalArticle

Makrogiannis, S, Fishbein, KW, Moore, AZ, Spencer, RG & Ferrucci, L 2016, 'Image-based tissue distribution modeling for skeletal muscle quality characterization', IEEE Transactions on Biomedical Engineering, vol. 63, no. 4, 7229301, pp. 805-813. https://doi.org/10.1109/TBME.2015.2474305
Makrogiannis, S. ; Fishbein, K. W. ; Moore, A. Z. ; Spencer, R. G. ; Ferrucci, L. / Image-based tissue distribution modeling for skeletal muscle quality characterization. In: IEEE Transactions on Biomedical Engineering. 2016 ; Vol. 63, No. 4. pp. 805-813.
@article{4fadb34ffbf344839fb7970ee3a9d28c,
title = "Image-based tissue distribution modeling for skeletal muscle quality characterization",
abstract = "The identification and characterization of regional body tissues is essential to understand changes that occur with aging and age-related metabolic diseases such as diabetes and obesity and how these diseases affect trajectories of health and functional status. Imaging technologies are frequently used to derive volumetric, area, and density measurements of different tissues. Despite the significance and direct applicability of automated tissue quantification and characterization techniques, these topics have remained relatively underexplored in the medical image analysis literature. We present a method for identification and characterization of muscle and adipose tissue in the midthigh region using MRI. We propose an image-based muscle quality prediction technique that estimates tissue-specific probability density models and their eigenstructures in the joint domain of water- and fat-suppressed voxel signal intensities along with volumetric and intensity-based tissue characteristics computed during the quantification stage. We evaluated the predictive capability of our approach against reference biomechanical muscle quality (MQ) measurements using statistical tests and classification performance experiments. The reference standard for MQ is defined as the ratio of muscle strength to muscle mass. The results show promise for the development of noninvasive image-based MQ descriptors.",
keywords = "Magnetic resonance imaging (MRI), Probabilistic modeling, Tissue identification and characterization",
author = "S. Makrogiannis and Fishbein, {K. W.} and Moore, {A. Z.} and Spencer, {R. G.} and L. Ferrucci",
year = "2016",
month = "4",
day = "1",
doi = "10.1109/TBME.2015.2474305",
language = "English (US)",
volume = "63",
pages = "805--813",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "4",

}

TY - JOUR

T1 - Image-based tissue distribution modeling for skeletal muscle quality characterization

AU - Makrogiannis, S.

AU - Fishbein, K. W.

AU - Moore, A. Z.

AU - Spencer, R. G.

AU - Ferrucci, L.

PY - 2016/4/1

Y1 - 2016/4/1

N2 - The identification and characterization of regional body tissues is essential to understand changes that occur with aging and age-related metabolic diseases such as diabetes and obesity and how these diseases affect trajectories of health and functional status. Imaging technologies are frequently used to derive volumetric, area, and density measurements of different tissues. Despite the significance and direct applicability of automated tissue quantification and characterization techniques, these topics have remained relatively underexplored in the medical image analysis literature. We present a method for identification and characterization of muscle and adipose tissue in the midthigh region using MRI. We propose an image-based muscle quality prediction technique that estimates tissue-specific probability density models and their eigenstructures in the joint domain of water- and fat-suppressed voxel signal intensities along with volumetric and intensity-based tissue characteristics computed during the quantification stage. We evaluated the predictive capability of our approach against reference biomechanical muscle quality (MQ) measurements using statistical tests and classification performance experiments. The reference standard for MQ is defined as the ratio of muscle strength to muscle mass. The results show promise for the development of noninvasive image-based MQ descriptors.

AB - The identification and characterization of regional body tissues is essential to understand changes that occur with aging and age-related metabolic diseases such as diabetes and obesity and how these diseases affect trajectories of health and functional status. Imaging technologies are frequently used to derive volumetric, area, and density measurements of different tissues. Despite the significance and direct applicability of automated tissue quantification and characterization techniques, these topics have remained relatively underexplored in the medical image analysis literature. We present a method for identification and characterization of muscle and adipose tissue in the midthigh region using MRI. We propose an image-based muscle quality prediction technique that estimates tissue-specific probability density models and their eigenstructures in the joint domain of water- and fat-suppressed voxel signal intensities along with volumetric and intensity-based tissue characteristics computed during the quantification stage. We evaluated the predictive capability of our approach against reference biomechanical muscle quality (MQ) measurements using statistical tests and classification performance experiments. The reference standard for MQ is defined as the ratio of muscle strength to muscle mass. The results show promise for the development of noninvasive image-based MQ descriptors.

KW - Magnetic resonance imaging (MRI)

KW - Probabilistic modeling

KW - Tissue identification and characterization

UR - http://www.scopus.com/inward/record.url?scp=84963558359&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963558359&partnerID=8YFLogxK

U2 - 10.1109/TBME.2015.2474305

DO - 10.1109/TBME.2015.2474305

M3 - Article

VL - 63

SP - 805

EP - 813

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 4

M1 - 7229301

ER -