Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy

Aydin Eresen, Noor E. Hafsa, Lejla Alic, Sharla M. Birch, John F. Griffin, Joe N. Kornegay, Jim Ji

Research output: Contribution to journalArticle

Abstract

Introduction: Golden retriever muscular dystrophy (GRMD) is a spontaneous X-linked canine model of Duchenne muscular dystrophy that resembles the human condition. Muscle percentage index (MPI) is proposed as an imaging biomarker of disease severity in GRMD. Methods: To assess MPI, we used MRI data acquired from nine GRMD samples using a 4.7 T small-bore scanner. A machine learning approach was used with eight raw quantitative mapping of MRI data images (T1m, T2m, two Dixon maps, and four diffusion tensor imaging maps), three types of texture descriptors (local binary pattern, gray-level co-occurrence matrix, gray-level run-length matrix), and a gradient descriptor (histogram of oriented gradients). Results: The confusion matrix, averaged over all samples, showed 93.5% of muscle pixels classified correctly. The classification, optimized in a leave-one-out cross-validation, provided an average accuracy of 80% with a discrepancy in overestimation for young (8%) and old (20%) dogs. Discussion: MPI could be useful for quantifying GRMD severity, but careful interpretation is needed for severe cases.

Original languageEnglish
JournalMuscle and Nerve
DOIs
Publication statusAccepted/In press - 1 Jan 2019

Fingerprint

Muscular Dystrophies
Muscles
Diffusion Tensor Imaging
Duchenne Muscular Dystrophy
Canidae
Biomarkers
Dogs

Keywords

  • DMD
  • GRMD
  • imaging biomarkers
  • machine learning
  • muscle percentage index
  • texture

ASJC Scopus subject areas

  • Physiology
  • Clinical Neurology
  • Cellular and Molecular Neuroscience
  • Physiology (medical)

Cite this

Eresen, A., Hafsa, N. E., Alic, L., Birch, S. M., Griffin, J. F., Kornegay, J. N., & Ji, J. (Accepted/In press). Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy. Muscle and Nerve. https://doi.org/10.1002/mus.26657

Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy. / Eresen, Aydin; Hafsa, Noor E.; Alic, Lejla; Birch, Sharla M.; Griffin, John F.; Kornegay, Joe N.; Ji, Jim.

In: Muscle and Nerve, 01.01.2019.

Research output: Contribution to journalArticle

Eresen, Aydin ; Hafsa, Noor E. ; Alic, Lejla ; Birch, Sharla M. ; Griffin, John F. ; Kornegay, Joe N. ; Ji, Jim. / Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy. In: Muscle and Nerve. 2019.
@article{db3e267fdf70489590e21bcb2f35e90c,
title = "Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy",
abstract = "Introduction: Golden retriever muscular dystrophy (GRMD) is a spontaneous X-linked canine model of Duchenne muscular dystrophy that resembles the human condition. Muscle percentage index (MPI) is proposed as an imaging biomarker of disease severity in GRMD. Methods: To assess MPI, we used MRI data acquired from nine GRMD samples using a 4.7 T small-bore scanner. A machine learning approach was used with eight raw quantitative mapping of MRI data images (T1m, T2m, two Dixon maps, and four diffusion tensor imaging maps), three types of texture descriptors (local binary pattern, gray-level co-occurrence matrix, gray-level run-length matrix), and a gradient descriptor (histogram of oriented gradients). Results: The confusion matrix, averaged over all samples, showed 93.5{\%} of muscle pixels classified correctly. The classification, optimized in a leave-one-out cross-validation, provided an average accuracy of 80{\%} with a discrepancy in overestimation for young (8{\%}) and old (20{\%}) dogs. Discussion: MPI could be useful for quantifying GRMD severity, but careful interpretation is needed for severe cases.",
keywords = "DMD, GRMD, imaging biomarkers, machine learning, muscle percentage index, texture",
author = "Aydin Eresen and Hafsa, {Noor E.} and Lejla Alic and Birch, {Sharla M.} and Griffin, {John F.} and Kornegay, {Joe N.} and Jim Ji",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/mus.26657",
language = "English",
journal = "Muscle and Nerve",
issn = "0148-639X",
publisher = "John Wiley and Sons Inc.",

}

TY - JOUR

T1 - Muscle percentage index as a marker of disease severity in golden retriever muscular dystrophy

AU - Eresen, Aydin

AU - Hafsa, Noor E.

AU - Alic, Lejla

AU - Birch, Sharla M.

AU - Griffin, John F.

AU - Kornegay, Joe N.

AU - Ji, Jim

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Introduction: Golden retriever muscular dystrophy (GRMD) is a spontaneous X-linked canine model of Duchenne muscular dystrophy that resembles the human condition. Muscle percentage index (MPI) is proposed as an imaging biomarker of disease severity in GRMD. Methods: To assess MPI, we used MRI data acquired from nine GRMD samples using a 4.7 T small-bore scanner. A machine learning approach was used with eight raw quantitative mapping of MRI data images (T1m, T2m, two Dixon maps, and four diffusion tensor imaging maps), three types of texture descriptors (local binary pattern, gray-level co-occurrence matrix, gray-level run-length matrix), and a gradient descriptor (histogram of oriented gradients). Results: The confusion matrix, averaged over all samples, showed 93.5% of muscle pixels classified correctly. The classification, optimized in a leave-one-out cross-validation, provided an average accuracy of 80% with a discrepancy in overestimation for young (8%) and old (20%) dogs. Discussion: MPI could be useful for quantifying GRMD severity, but careful interpretation is needed for severe cases.

AB - Introduction: Golden retriever muscular dystrophy (GRMD) is a spontaneous X-linked canine model of Duchenne muscular dystrophy that resembles the human condition. Muscle percentage index (MPI) is proposed as an imaging biomarker of disease severity in GRMD. Methods: To assess MPI, we used MRI data acquired from nine GRMD samples using a 4.7 T small-bore scanner. A machine learning approach was used with eight raw quantitative mapping of MRI data images (T1m, T2m, two Dixon maps, and four diffusion tensor imaging maps), three types of texture descriptors (local binary pattern, gray-level co-occurrence matrix, gray-level run-length matrix), and a gradient descriptor (histogram of oriented gradients). Results: The confusion matrix, averaged over all samples, showed 93.5% of muscle pixels classified correctly. The classification, optimized in a leave-one-out cross-validation, provided an average accuracy of 80% with a discrepancy in overestimation for young (8%) and old (20%) dogs. Discussion: MPI could be useful for quantifying GRMD severity, but careful interpretation is needed for severe cases.

KW - DMD

KW - GRMD

KW - imaging biomarkers

KW - machine learning

KW - muscle percentage index

KW - texture

UR - http://www.scopus.com/inward/record.url?scp=85071651518&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071651518&partnerID=8YFLogxK

U2 - 10.1002/mus.26657

DO - 10.1002/mus.26657

M3 - Article

C2 - 31397906

AN - SCOPUS:85071651518

JO - Muscle and Nerve

JF - Muscle and Nerve

SN - 0148-639X

ER -