Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation

Angel Torrado-Carvajal, Yigitcan Eryaman, Esra Abaci Turk, Joaquin L. Herraiz, Juan A. Hernandez-Tamames, Elfar Adalsteinsson, Lawrence L. Wald, Norberto Malpica

Research output: Contribution to journalArticle

Abstract

Objective: The purpose of this paper is to prove that computer-vision techniques allow synthesizing water-fat separation maps for local specific absorption rate (SAR) estimation, when patient-specific water-fat images are not available. Methods: We obtained ground truth head models by using patient-specific water-fat images. We obtained two different label-fusion water-fat models generating a water-fat multiatlas and applying the STAPLE and local-MAP-STAPLE label-fusion methods. We also obtained patch-based water-fat models applying a local group-wise weighted combination of the multiatlas. Electromagnetic (EM) simulations were performed, and B1+ magnitude and 10 g averaged SAR maps were generated. Results: We found local approaches provide a high DICE overlap (72.6 ± 10.2% fat and 91.6 ± 1.5% water in local-MAP-STAPLE, and 68.8 ± 8.2% fat and 91.1 ± 1.0% water in patch-based), low Hausdorff distances (18.6 ± 7.7 mm fat and 7.4 ± 11.2 mm water in local-MAP-STAPLE, and 16.4 ± 8.5 mm fat and 7.2 ± 11.8 mm water in patch-based) and a low error in volume estimation (15.6 ± 34.4% fat and 5.6 ± 4.1% water in the local-MAP-STAPLE, and 14.0 ± 17.7% fat and 4.7 ± 2.8% water in patch-based). The positions of the peak 10 g-averaged local SAR hotspots were the same for every model. Conclusion: We have created patient-specific head models using three different computer-vision-based water-fat separation approaches and compared the predictions of B1+ field and SAR distributions generated by simulating these models. Our results prove that a computer-vision approach can be used for patient-specific water-fat separation, and utilized for local SAR estimation in high-field MRI. Significance: Computer-vision approaches can be used for patient-specific water-fat separation and for patient specific local SAR estimation, when water-fat images of the patient are not available.

Original languageEnglish (US)
Article number8411472
Pages (from-to)768-774
Number of pages7
JournalIEEE Transactions on Biomedical Engineering
Volume66
Issue number3
DOIs
StatePublished - Mar 1 2019

Fingerprint

Oils and fats
Magnetic resonance imaging
Computer vision
Water
Labels
Fusion reactions

Keywords

  • Computer vision
  • MRI
  • SAR management
  • head models
  • magnetic resonance imaging
  • multi-atlas segmentation
  • water-fat separation

PubMed: MeSH publication types

  • Journal Article
  • Research Support, Non-U.S. Gov't

Cite this

Torrado-Carvajal, A., Eryaman, Y., Turk, E. A., Herraiz, J. L., Hernandez-Tamames, J. A., Adalsteinsson, E., ... Malpica, N. (2019). Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation. IEEE Transactions on Biomedical Engineering, 66(3), 768-774. [8411472]. https://doi.org/10.1109/TBME.2018.2856501

Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation. / Torrado-Carvajal, Angel; Eryaman, Yigitcan; Turk, Esra Abaci; Herraiz, Joaquin L.; Hernandez-Tamames, Juan A.; Adalsteinsson, Elfar; Wald, Lawrence L.; Malpica, Norberto.

In: IEEE Transactions on Biomedical Engineering, Vol. 66, No. 3, 8411472, 01.03.2019, p. 768-774.

Research output: Contribution to journalArticle

Torrado-Carvajal, A, Eryaman, Y, Turk, EA, Herraiz, JL, Hernandez-Tamames, JA, Adalsteinsson, E, Wald, LL & Malpica, N 2019, 'Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation', IEEE Transactions on Biomedical Engineering, vol. 66, no. 3, 8411472, pp. 768-774. https://doi.org/10.1109/TBME.2018.2856501
Torrado-Carvajal, Angel ; Eryaman, Yigitcan ; Turk, Esra Abaci ; Herraiz, Joaquin L. ; Hernandez-Tamames, Juan A. ; Adalsteinsson, Elfar ; Wald, Lawrence L. ; Malpica, Norberto. / Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation. In: IEEE Transactions on Biomedical Engineering. 2019 ; Vol. 66, No. 3. pp. 768-774.
@article{ee61419914fb4a5796b2a51ccbecccfb,
title = "Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation",
abstract = "Objective: The purpose of this paper is to prove that computer-vision techniques allow synthesizing water-fat separation maps for local specific absorption rate (SAR) estimation, when patient-specific water-fat images are not available. Methods: We obtained ground truth head models by using patient-specific water-fat images. We obtained two different label-fusion water-fat models generating a water-fat multiatlas and applying the STAPLE and local-MAP-STAPLE label-fusion methods. We also obtained patch-based water-fat models applying a local group-wise weighted combination of the multiatlas. Electromagnetic (EM) simulations were performed, and B1+ magnitude and 10 g averaged SAR maps were generated. Results: We found local approaches provide a high DICE overlap (72.6 ± 10.2{\%} fat and 91.6 ± 1.5{\%} water in local-MAP-STAPLE, and 68.8 ± 8.2{\%} fat and 91.1 ± 1.0{\%} water in patch-based), low Hausdorff distances (18.6 ± 7.7 mm fat and 7.4 ± 11.2 mm water in local-MAP-STAPLE, and 16.4 ± 8.5 mm fat and 7.2 ± 11.8 mm water in patch-based) and a low error in volume estimation (15.6 ± 34.4{\%} fat and 5.6 ± 4.1{\%} water in the local-MAP-STAPLE, and 14.0 ± 17.7{\%} fat and 4.7 ± 2.8{\%} water in patch-based). The positions of the peak 10 g-averaged local SAR hotspots were the same for every model. Conclusion: We have created patient-specific head models using three different computer-vision-based water-fat separation approaches and compared the predictions of B1+ field and SAR distributions generated by simulating these models. Our results prove that a computer-vision approach can be used for patient-specific water-fat separation, and utilized for local SAR estimation in high-field MRI. Significance: Computer-vision approaches can be used for patient-specific water-fat separation and for patient specific local SAR estimation, when water-fat images of the patient are not available.",
keywords = "Computer vision, MRI, SAR management, head models, magnetic resonance imaging, multi-atlas segmentation, water-fat separation",
author = "Angel Torrado-Carvajal and Yigitcan Eryaman and Turk, {Esra Abaci} and Herraiz, {Joaquin L.} and Hernandez-Tamames, {Juan A.} and Elfar Adalsteinsson and Wald, {Lawrence L.} and Norberto Malpica",
year = "2019",
month = "3",
day = "1",
doi = "10.1109/TBME.2018.2856501",
language = "English (US)",
volume = "66",
pages = "768--774",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "3",

}

TY - JOUR

T1 - Computer-Vision Techniques for Water-Fat Separation in Ultra High-Field MRI Local Specific Absorption Rate Estimation

AU - Torrado-Carvajal, Angel

AU - Eryaman, Yigitcan

AU - Turk, Esra Abaci

AU - Herraiz, Joaquin L.

AU - Hernandez-Tamames, Juan A.

AU - Adalsteinsson, Elfar

AU - Wald, Lawrence L.

AU - Malpica, Norberto

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Objective: The purpose of this paper is to prove that computer-vision techniques allow synthesizing water-fat separation maps for local specific absorption rate (SAR) estimation, when patient-specific water-fat images are not available. Methods: We obtained ground truth head models by using patient-specific water-fat images. We obtained two different label-fusion water-fat models generating a water-fat multiatlas and applying the STAPLE and local-MAP-STAPLE label-fusion methods. We also obtained patch-based water-fat models applying a local group-wise weighted combination of the multiatlas. Electromagnetic (EM) simulations were performed, and B1+ magnitude and 10 g averaged SAR maps were generated. Results: We found local approaches provide a high DICE overlap (72.6 ± 10.2% fat and 91.6 ± 1.5% water in local-MAP-STAPLE, and 68.8 ± 8.2% fat and 91.1 ± 1.0% water in patch-based), low Hausdorff distances (18.6 ± 7.7 mm fat and 7.4 ± 11.2 mm water in local-MAP-STAPLE, and 16.4 ± 8.5 mm fat and 7.2 ± 11.8 mm water in patch-based) and a low error in volume estimation (15.6 ± 34.4% fat and 5.6 ± 4.1% water in the local-MAP-STAPLE, and 14.0 ± 17.7% fat and 4.7 ± 2.8% water in patch-based). The positions of the peak 10 g-averaged local SAR hotspots were the same for every model. Conclusion: We have created patient-specific head models using three different computer-vision-based water-fat separation approaches and compared the predictions of B1+ field and SAR distributions generated by simulating these models. Our results prove that a computer-vision approach can be used for patient-specific water-fat separation, and utilized for local SAR estimation in high-field MRI. Significance: Computer-vision approaches can be used for patient-specific water-fat separation and for patient specific local SAR estimation, when water-fat images of the patient are not available.

AB - Objective: The purpose of this paper is to prove that computer-vision techniques allow synthesizing water-fat separation maps for local specific absorption rate (SAR) estimation, when patient-specific water-fat images are not available. Methods: We obtained ground truth head models by using patient-specific water-fat images. We obtained two different label-fusion water-fat models generating a water-fat multiatlas and applying the STAPLE and local-MAP-STAPLE label-fusion methods. We also obtained patch-based water-fat models applying a local group-wise weighted combination of the multiatlas. Electromagnetic (EM) simulations were performed, and B1+ magnitude and 10 g averaged SAR maps were generated. Results: We found local approaches provide a high DICE overlap (72.6 ± 10.2% fat and 91.6 ± 1.5% water in local-MAP-STAPLE, and 68.8 ± 8.2% fat and 91.1 ± 1.0% water in patch-based), low Hausdorff distances (18.6 ± 7.7 mm fat and 7.4 ± 11.2 mm water in local-MAP-STAPLE, and 16.4 ± 8.5 mm fat and 7.2 ± 11.8 mm water in patch-based) and a low error in volume estimation (15.6 ± 34.4% fat and 5.6 ± 4.1% water in the local-MAP-STAPLE, and 14.0 ± 17.7% fat and 4.7 ± 2.8% water in patch-based). The positions of the peak 10 g-averaged local SAR hotspots were the same for every model. Conclusion: We have created patient-specific head models using three different computer-vision-based water-fat separation approaches and compared the predictions of B1+ field and SAR distributions generated by simulating these models. Our results prove that a computer-vision approach can be used for patient-specific water-fat separation, and utilized for local SAR estimation in high-field MRI. Significance: Computer-vision approaches can be used for patient-specific water-fat separation and for patient specific local SAR estimation, when water-fat images of the patient are not available.

KW - Computer vision

KW - MRI

KW - SAR management

KW - head models

KW - magnetic resonance imaging

KW - multi-atlas segmentation

KW - water-fat separation

UR - http://www.scopus.com/inward/record.url?scp=85049955612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049955612&partnerID=8YFLogxK

U2 - 10.1109/TBME.2018.2856501

DO - 10.1109/TBME.2018.2856501

M3 - Article

VL - 66

SP - 768

EP - 774

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 3

M1 - 8411472

ER -