A Sparse Non-negative Matrix Factorization Framework for Identifying Functional Units of Tongue Behavior from MRI

Jonghye Woo, Jerry Ladd Prince, Maureen Stone, Fangxu Xing, Arnold Gomez, Jordan R. Green, Christopher J. Hartnick, Thomas J. Brady, Timothy G. Reese, Van J. Wedeen, Georges El Fakhri

Research output: Contribution to journalArticle

Abstract

Muscle coordination patterns of lingual behaviors are synergies generated by deforming local muscle groups in a variety of ways. Functional units are functional muscle groups of local structural elements within the tongue that compress, expand, and move in a cohesive and consistent manner. Identifying the functional units using tagged-Magnetic Resonance Imaging (MRI) sheds light on the mechanisms of normal and pathological muscle coordination patterns, yielding improvement in surgical planning, treatment, or rehabilitation procedures. Here, to mine this information, we propose a matrix factorization and probabilistic graphical model framework to produce building blocks and their associated weighting map using motion quantities extracted from tagged-MRI. Our tagged-MRI imaging and accurate voxel-level tracking provide previously unavailable internal tongue motion patterns, thus revealing the inner workings of the tongue during speech or other lingual behaviors. We then employ spectral clustering on the weighting map to identify the cohesive regions defined by the tongue motion that may involve multiple or undocumented regions. To evaluate our method, we perform a series of experiments. We first use two-dimensional images and synthetic data to demonstrate the accuracy of our method. We then use three-dimensional synthetic and in vivo tongue motion data using protrusion and simple speech tasks to identify subject-specific and data-driven functional units of the tongue in localized regions.

Original languageEnglish (US)
JournalIEEE Transactions on Medical Imaging
DOIs
StateAccepted/In press - Sep 17 2018

Fingerprint

Factorization
Tongue
Muscle
Magnetic Resonance Imaging
Muscles
Patient rehabilitation
Imaging techniques
Planning
Statistical Models
Cluster Analysis
Rehabilitation
Experiments

Keywords

  • Biomedical imaging
  • Functional Units
  • Magnetic resonance imaging
  • MRI
  • Muscles
  • Nonnegative Matrix Factorization
  • Sparse matrices
  • Sparsity
  • Speech
  • Task analysis
  • Tongue
  • Tongue Motion
  • Tracking

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

A Sparse Non-negative Matrix Factorization Framework for Identifying Functional Units of Tongue Behavior from MRI. / Woo, Jonghye; Prince, Jerry Ladd; Stone, Maureen; Xing, Fangxu; Gomez, Arnold; Green, Jordan R.; Hartnick, Christopher J.; Brady, Thomas J.; Reese, Timothy G.; Wedeen, Van J.; El Fakhri, Georges.

In: IEEE Transactions on Medical Imaging, 17.09.2018.

Research output: Contribution to journalArticle

Woo, Jonghye ; Prince, Jerry Ladd ; Stone, Maureen ; Xing, Fangxu ; Gomez, Arnold ; Green, Jordan R. ; Hartnick, Christopher J. ; Brady, Thomas J. ; Reese, Timothy G. ; Wedeen, Van J. ; El Fakhri, Georges. / A Sparse Non-negative Matrix Factorization Framework for Identifying Functional Units of Tongue Behavior from MRI. In: IEEE Transactions on Medical Imaging. 2018.
@article{b8181368fc2349a888f701121d8691c0,
title = "A Sparse Non-negative Matrix Factorization Framework for Identifying Functional Units of Tongue Behavior from MRI",
abstract = "Muscle coordination patterns of lingual behaviors are synergies generated by deforming local muscle groups in a variety of ways. Functional units are functional muscle groups of local structural elements within the tongue that compress, expand, and move in a cohesive and consistent manner. Identifying the functional units using tagged-Magnetic Resonance Imaging (MRI) sheds light on the mechanisms of normal and pathological muscle coordination patterns, yielding improvement in surgical planning, treatment, or rehabilitation procedures. Here, to mine this information, we propose a matrix factorization and probabilistic graphical model framework to produce building blocks and their associated weighting map using motion quantities extracted from tagged-MRI. Our tagged-MRI imaging and accurate voxel-level tracking provide previously unavailable internal tongue motion patterns, thus revealing the inner workings of the tongue during speech or other lingual behaviors. We then employ spectral clustering on the weighting map to identify the cohesive regions defined by the tongue motion that may involve multiple or undocumented regions. To evaluate our method, we perform a series of experiments. We first use two-dimensional images and synthetic data to demonstrate the accuracy of our method. We then use three-dimensional synthetic and in vivo tongue motion data using protrusion and simple speech tasks to identify subject-specific and data-driven functional units of the tongue in localized regions.",
keywords = "Biomedical imaging, Functional Units, Magnetic resonance imaging, MRI, Muscles, Nonnegative Matrix Factorization, Sparse matrices, Sparsity, Speech, Task analysis, Tongue, Tongue Motion, Tracking",
author = "Jonghye Woo and Prince, {Jerry Ladd} and Maureen Stone and Fangxu Xing and Arnold Gomez and Green, {Jordan R.} and Hartnick, {Christopher J.} and Brady, {Thomas J.} and Reese, {Timothy G.} and Wedeen, {Van J.} and {El Fakhri}, Georges",
year = "2018",
month = "9",
day = "17",
doi = "10.1109/TMI.2018.2870939",
language = "English (US)",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - A Sparse Non-negative Matrix Factorization Framework for Identifying Functional Units of Tongue Behavior from MRI

AU - Woo, Jonghye

AU - Prince, Jerry Ladd

AU - Stone, Maureen

AU - Xing, Fangxu

AU - Gomez, Arnold

AU - Green, Jordan R.

AU - Hartnick, Christopher J.

AU - Brady, Thomas J.

AU - Reese, Timothy G.

AU - Wedeen, Van J.

AU - El Fakhri, Georges

PY - 2018/9/17

Y1 - 2018/9/17

N2 - Muscle coordination patterns of lingual behaviors are synergies generated by deforming local muscle groups in a variety of ways. Functional units are functional muscle groups of local structural elements within the tongue that compress, expand, and move in a cohesive and consistent manner. Identifying the functional units using tagged-Magnetic Resonance Imaging (MRI) sheds light on the mechanisms of normal and pathological muscle coordination patterns, yielding improvement in surgical planning, treatment, or rehabilitation procedures. Here, to mine this information, we propose a matrix factorization and probabilistic graphical model framework to produce building blocks and their associated weighting map using motion quantities extracted from tagged-MRI. Our tagged-MRI imaging and accurate voxel-level tracking provide previously unavailable internal tongue motion patterns, thus revealing the inner workings of the tongue during speech or other lingual behaviors. We then employ spectral clustering on the weighting map to identify the cohesive regions defined by the tongue motion that may involve multiple or undocumented regions. To evaluate our method, we perform a series of experiments. We first use two-dimensional images and synthetic data to demonstrate the accuracy of our method. We then use three-dimensional synthetic and in vivo tongue motion data using protrusion and simple speech tasks to identify subject-specific and data-driven functional units of the tongue in localized regions.

AB - Muscle coordination patterns of lingual behaviors are synergies generated by deforming local muscle groups in a variety of ways. Functional units are functional muscle groups of local structural elements within the tongue that compress, expand, and move in a cohesive and consistent manner. Identifying the functional units using tagged-Magnetic Resonance Imaging (MRI) sheds light on the mechanisms of normal and pathological muscle coordination patterns, yielding improvement in surgical planning, treatment, or rehabilitation procedures. Here, to mine this information, we propose a matrix factorization and probabilistic graphical model framework to produce building blocks and their associated weighting map using motion quantities extracted from tagged-MRI. Our tagged-MRI imaging and accurate voxel-level tracking provide previously unavailable internal tongue motion patterns, thus revealing the inner workings of the tongue during speech or other lingual behaviors. We then employ spectral clustering on the weighting map to identify the cohesive regions defined by the tongue motion that may involve multiple or undocumented regions. To evaluate our method, we perform a series of experiments. We first use two-dimensional images and synthetic data to demonstrate the accuracy of our method. We then use three-dimensional synthetic and in vivo tongue motion data using protrusion and simple speech tasks to identify subject-specific and data-driven functional units of the tongue in localized regions.

KW - Biomedical imaging

KW - Functional Units

KW - Magnetic resonance imaging

KW - MRI

KW - Muscles

KW - Nonnegative Matrix Factorization

KW - Sparse matrices

KW - Sparsity

KW - Speech

KW - Task analysis

KW - Tongue

KW - Tongue Motion

KW - Tracking

UR - http://www.scopus.com/inward/record.url?scp=85053619241&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053619241&partnerID=8YFLogxK

U2 - 10.1109/TMI.2018.2870939

DO - 10.1109/TMI.2018.2870939

M3 - Article

C2 - 30235120

AN - SCOPUS:85053619241

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

ER -