Reproducibility assessment of ultrasound-based aortic stiffness quantification and verification using Bi-axial tensile testing

Emiel M.J. van Disseldorp, Marcel H.M.H. van den Hoven, Frans N. van de Vosse, Marc R.H.M. van Sambeek, Richard G.P. Lopata (Corresponding author)

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

Uittreksel

Current guidelines for abdominal aortic aneurysm (AAA) repair are primarily based on the maximum diameter. Since these methods lack robustness in decision making, new image-based methods for mechanical characterization have been proposed. Recently, time-resolved 3D ultrasound (4D US) in combination with finite element analysis was shown to provide additional risk estimators such as patient-specific peak wall stresses and wall stiffness in a non-invasive way. The aim of this study is to: 1) assess the reproducibility of this US-based stiffness measurement in vitro and in vivo, and 2) verify this 4D US stiffness using the gold standard: bi-axial tensile testing of the excised aortic tissue. For the in vitro study, 4D US data were acquired in an idealized inflation experiment using porcine aortas. The full aortic geometry was segmented and tracked over the cardiac cycle, and afterwards finite element analysis was performed by calibrating the finite element model to the measured US displacements to find the global aortic wall stiffness. For verification purposes, the porcine tissue was subjected to bi-axial tensile testing. Secondly, four AAA patients were included and 4D US data were acquired before open aortic surgery was performed. Similar to the experimental approach, the 4D US data were analyzed using the iterative finite element approach. During surgery, aortic tissue was harvested and the resulting tissue specimens were analyzed using bi-axial tensile testing. Finally, reproducibility was quantified for both methods. A high reproducibility was observed for the wall stiffness measurements using 4D US, i.e., an ICC of 0.91 (95% CI: 0.78–0.98) for the porcine aortas and an ICC of 0.98 (95% CI: 0.84–1.00) for the AAA samples. Verification with bi-axial tensile testing revealed a good agreement for the inflation experiment and a moderate agreement for the AAA patients, partially caused by the diseased state and inhomogeneities of the tissue. The performance of aortic stiffness characterization using 4D US revealed overall a high reproducibility and a moderate agreement with ex vivo mechanical testing. Future research should include more patient samples, to statistically assess the accuracy of the current in vivo method, which is not trivial due to the low number of open surgical interventions.

Originele taal-2Engels
Artikelnummer103571
Aantal pagina's10
TijdschriftJournal of the Mechanical Behavior of Biomedical Materials
Volume103
DOI's
StatusGepubliceerd - mrt 2020

Vingerafdruk

Tensile testing
Ultrasonics
Stiffness
Tissue
Surgery
Finite element method
Mechanical testing
Repair
Decision making
Experiments
Geometry

Citeer dit

@article{4bd2d582520e4ec18d75b34d05305834,
title = "Reproducibility assessment of ultrasound-based aortic stiffness quantification and verification using Bi-axial tensile testing",
abstract = "Current guidelines for abdominal aortic aneurysm (AAA) repair are primarily based on the maximum diameter. Since these methods lack robustness in decision making, new image-based methods for mechanical characterization have been proposed. Recently, time-resolved 3D ultrasound (4D US) in combination with finite element analysis was shown to provide additional risk estimators such as patient-specific peak wall stresses and wall stiffness in a non-invasive way. The aim of this study is to: 1) assess the reproducibility of this US-based stiffness measurement in vitro and in vivo, and 2) verify this 4D US stiffness using the gold standard: bi-axial tensile testing of the excised aortic tissue. For the in vitro study, 4D US data were acquired in an idealized inflation experiment using porcine aortas. The full aortic geometry was segmented and tracked over the cardiac cycle, and afterwards finite element analysis was performed by calibrating the finite element model to the measured US displacements to find the global aortic wall stiffness. For verification purposes, the porcine tissue was subjected to bi-axial tensile testing. Secondly, four AAA patients were included and 4D US data were acquired before open aortic surgery was performed. Similar to the experimental approach, the 4D US data were analyzed using the iterative finite element approach. During surgery, aortic tissue was harvested and the resulting tissue specimens were analyzed using bi-axial tensile testing. Finally, reproducibility was quantified for both methods. A high reproducibility was observed for the wall stiffness measurements using 4D US, i.e., an ICC of 0.91 (95{\%} CI: 0.78–0.98) for the porcine aortas and an ICC of 0.98 (95{\%} CI: 0.84–1.00) for the AAA samples. Verification with bi-axial tensile testing revealed a good agreement for the inflation experiment and a moderate agreement for the AAA patients, partially caused by the diseased state and inhomogeneities of the tissue. The performance of aortic stiffness characterization using 4D US revealed overall a high reproducibility and a moderate agreement with ex vivo mechanical testing. Future research should include more patient samples, to statistically assess the accuracy of the current in vivo method, which is not trivial due to the low number of open surgical interventions.",
keywords = "3D ultrasound, Abdominal aortic aneurysm, Reproducibility, Verification",
author = "{van Disseldorp}, {Emiel M.J.} and {van den Hoven}, {Marcel H.M.H.} and {van de Vosse}, {Frans N.} and {van Sambeek}, {Marc R.H.M.} and Lopata, {Richard G.P.}",
year = "2020",
month = "3",
doi = "10.1016/j.jmbbm.2019.103571",
language = "English",
volume = "103",
journal = "Journal of the Mechanical Behavior of Biomedical Materials",
issn = "1751-6161",
publisher = "Elsevier",

}

Reproducibility assessment of ultrasound-based aortic stiffness quantification and verification using Bi-axial tensile testing. / van Disseldorp, Emiel M.J.; van den Hoven, Marcel H.M.H.; van de Vosse, Frans N.; van Sambeek, Marc R.H.M.; Lopata, Richard G.P. (Corresponding author).

In: Journal of the Mechanical Behavior of Biomedical Materials, Vol. 103, 103571, 03.2020.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Reproducibility assessment of ultrasound-based aortic stiffness quantification and verification using Bi-axial tensile testing

AU - van Disseldorp, Emiel M.J.

AU - van den Hoven, Marcel H.M.H.

AU - van de Vosse, Frans N.

AU - van Sambeek, Marc R.H.M.

AU - Lopata, Richard G.P.

PY - 2020/3

Y1 - 2020/3

N2 - Current guidelines for abdominal aortic aneurysm (AAA) repair are primarily based on the maximum diameter. Since these methods lack robustness in decision making, new image-based methods for mechanical characterization have been proposed. Recently, time-resolved 3D ultrasound (4D US) in combination with finite element analysis was shown to provide additional risk estimators such as patient-specific peak wall stresses and wall stiffness in a non-invasive way. The aim of this study is to: 1) assess the reproducibility of this US-based stiffness measurement in vitro and in vivo, and 2) verify this 4D US stiffness using the gold standard: bi-axial tensile testing of the excised aortic tissue. For the in vitro study, 4D US data were acquired in an idealized inflation experiment using porcine aortas. The full aortic geometry was segmented and tracked over the cardiac cycle, and afterwards finite element analysis was performed by calibrating the finite element model to the measured US displacements to find the global aortic wall stiffness. For verification purposes, the porcine tissue was subjected to bi-axial tensile testing. Secondly, four AAA patients were included and 4D US data were acquired before open aortic surgery was performed. Similar to the experimental approach, the 4D US data were analyzed using the iterative finite element approach. During surgery, aortic tissue was harvested and the resulting tissue specimens were analyzed using bi-axial tensile testing. Finally, reproducibility was quantified for both methods. A high reproducibility was observed for the wall stiffness measurements using 4D US, i.e., an ICC of 0.91 (95% CI: 0.78–0.98) for the porcine aortas and an ICC of 0.98 (95% CI: 0.84–1.00) for the AAA samples. Verification with bi-axial tensile testing revealed a good agreement for the inflation experiment and a moderate agreement for the AAA patients, partially caused by the diseased state and inhomogeneities of the tissue. The performance of aortic stiffness characterization using 4D US revealed overall a high reproducibility and a moderate agreement with ex vivo mechanical testing. Future research should include more patient samples, to statistically assess the accuracy of the current in vivo method, which is not trivial due to the low number of open surgical interventions.

AB - Current guidelines for abdominal aortic aneurysm (AAA) repair are primarily based on the maximum diameter. Since these methods lack robustness in decision making, new image-based methods for mechanical characterization have been proposed. Recently, time-resolved 3D ultrasound (4D US) in combination with finite element analysis was shown to provide additional risk estimators such as patient-specific peak wall stresses and wall stiffness in a non-invasive way. The aim of this study is to: 1) assess the reproducibility of this US-based stiffness measurement in vitro and in vivo, and 2) verify this 4D US stiffness using the gold standard: bi-axial tensile testing of the excised aortic tissue. For the in vitro study, 4D US data were acquired in an idealized inflation experiment using porcine aortas. The full aortic geometry was segmented and tracked over the cardiac cycle, and afterwards finite element analysis was performed by calibrating the finite element model to the measured US displacements to find the global aortic wall stiffness. For verification purposes, the porcine tissue was subjected to bi-axial tensile testing. Secondly, four AAA patients were included and 4D US data were acquired before open aortic surgery was performed. Similar to the experimental approach, the 4D US data were analyzed using the iterative finite element approach. During surgery, aortic tissue was harvested and the resulting tissue specimens were analyzed using bi-axial tensile testing. Finally, reproducibility was quantified for both methods. A high reproducibility was observed for the wall stiffness measurements using 4D US, i.e., an ICC of 0.91 (95% CI: 0.78–0.98) for the porcine aortas and an ICC of 0.98 (95% CI: 0.84–1.00) for the AAA samples. Verification with bi-axial tensile testing revealed a good agreement for the inflation experiment and a moderate agreement for the AAA patients, partially caused by the diseased state and inhomogeneities of the tissue. The performance of aortic stiffness characterization using 4D US revealed overall a high reproducibility and a moderate agreement with ex vivo mechanical testing. Future research should include more patient samples, to statistically assess the accuracy of the current in vivo method, which is not trivial due to the low number of open surgical interventions.

KW - 3D ultrasound

KW - Abdominal aortic aneurysm

KW - Reproducibility

KW - Verification

UR - http://www.scopus.com/inward/record.url?scp=85077443274&partnerID=8YFLogxK

U2 - 10.1016/j.jmbbm.2019.103571

DO - 10.1016/j.jmbbm.2019.103571

M3 - Article

AN - SCOPUS:85077443274

VL - 103

JO - Journal of the Mechanical Behavior of Biomedical Materials

JF - Journal of the Mechanical Behavior of Biomedical Materials

SN - 1751-6161

M1 - 103571

ER -