Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure

E. Ferdian; D. Marlevi; J. Schollenberger; M. Aristova; E. R. Edelman; S. Schnell; C. A. Figueroa; D. A. Nordsletten; A. A. Young

doi:10.1016/j.media.2023.102831

Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure

E. Ferdian^*, D. Marlevi, J. Schollenberger, M. Aristova, E. R. Edelman, S. Schnell, C. A. Figueroa, D. A. Nordsletten, A. A. Young

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

7 Scopus citations

Abstract

The development of cerebrovascular disease is tightly coupled to regional changes in intracranial flow and relative pressure. Image-based assessment using phase contrast magnetic resonance imaging has particular promise for non-invasive full-field mapping of cerebrovascular hemodynamics. However, estimations are complicated by the narrow and tortuous intracranial vasculature, with accurate image-based quantification directly dependent on sufficient spatial resolution. Further, extended scan times are required for high-resolution acquisitions, and most clinical acquisitions are performed at comparably low resolution (>1 mm) where biases have been observed with regard to the quantification of both flow and relative pressure. The aim of our study was to develop an approach for quantitative intracranial super-resolution 4D Flow MRI, with effective resolution enhancement achieved by a dedicated deep residual network, and with accurate quantification of functional relative pressures achieved by subsequent physics-informed image processing. To achieve this, our two-step approach was trained and validated in a patient-specific in-silico cohort, showing good accuracy in estimating velocity (relative error: 15.0 ± 0.1%, mean absolute error (MAE): 0.07 ± 0.06 m/s, and cosine similarity: 0.99 ± 0.06 at peak velocity) and flow (relative error: 6.6 ± 4.7%, root mean square error (RMSE): 0.56 mL/s at peak flow), and with the coupled physics-informed image analysis allowing for maintained recovery of functional relative pressure throughout the circle of Willis (relative error: 11.0 ± 7.3%, RMSE: 0.3 ± 0.2 mmHg). Furthermore, the quantitative super-resolution approach is applied to an in-vivo volunteer cohort, effectively generating intracranial flow images at <0.5 mm resolution and showing reduced low-resolution bias in relative pressure estimation. Our work thus presents a promising two-step approach to non-invasively quantify cerebrovascular hemodynamics, being applicable to dedicated clinical cohorts in the future.

Original language	English (US)
Article number	102831
Journal	Medical Image Analysis
Volume	88
DOIs	https://doi.org/10.1016/j.media.2023.102831
State	Published - Aug 2023

Funding

E. F. holds a New Zealand Heart Foundation Scholarship, Grant No. 1786. D.M. holds a Knut and Alice Wallenberg Foundation scholarship for postdoctoral studies at Massachusetts Institute of Technology. J.S. is supported by a University of Michigan Rackham Predoctoral Fellowship. M.A. was supported by a Ruth L. Kirschstein National Research Service Award (NIH F30 HL140910) and the Northwestern – Medical Science Training Program (NIH T32 GM815229). E.R.E. was funded in part by NIH R01 49039. A.A.Y. acknowledges core funding from the Wellcome/EPSRC Centre for Medical Engineering (WT203148/Z/16/Z) and the London Medical Imaging and AI Centre for Value-Based Healthcare. D.N. would like to acknowledge funding from the Engineering and Physical Science Research Council (EP/N011554 and EP/R003866/1). D.N. would like to acknowledge funding from the Engineering and Physical Science Research Council ( EP/N011554 and EP/R003866/1 ). A.A.Y. acknowledges core funding from the Wellcome/EPSRC Centre for Medical Engineering ( WT203148/Z/16/Z ) and the London Medical Imaging and AI Centre for Value-Based Healthcare. M.A. was supported by a Ruth L. Kirschstein National Research Service Award ( NIH F30 HL140910 ) and the Northwestern – Medical Science Training Program ( NIH T32 GM815229 ). E.R.E. was funded in part by NIH R01 49039 . J.S. is supported by a University of Michigan Rackham Predoctoral Fellowship . E. F. holds a New Zealand Heart Foundation Scholarship , Grant No. 1786 .

Keywords

Cerebrovasculature
Deep learning
Relative pressure
Super-resolution 4D flow MRI

ASJC Scopus subject areas

Radiological and Ultrasound Technology
Radiology Nuclear Medicine and imaging
Computer Vision and Pattern Recognition
Health Informatics
Computer Graphics and Computer-Aided Design

Access to Document

10.1016/j.media.2023.102831

Fingerprint

Dive into the research topics of 'Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure'. Together they form a unique fingerprint.

Cite this

Ferdian, E., Marlevi, D., Schollenberger, J., Aristova, M., Edelman, E. R., Schnell, S., Figueroa, C. A., Nordsletten, D. A., & Young, A. A. (2023). Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure. Medical Image Analysis, 88, Article 102831. https://doi.org/10.1016/j.media.2023.102831

@article{0dd5e5f038a04fbeb88a0f70ac52e762,

title = "Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure",

abstract = "The development of cerebrovascular disease is tightly coupled to regional changes in intracranial flow and relative pressure. Image-based assessment using phase contrast magnetic resonance imaging has particular promise for non-invasive full-field mapping of cerebrovascular hemodynamics. However, estimations are complicated by the narrow and tortuous intracranial vasculature, with accurate image-based quantification directly dependent on sufficient spatial resolution. Further, extended scan times are required for high-resolution acquisitions, and most clinical acquisitions are performed at comparably low resolution (>1 mm) where biases have been observed with regard to the quantification of both flow and relative pressure. The aim of our study was to develop an approach for quantitative intracranial super-resolution 4D Flow MRI, with effective resolution enhancement achieved by a dedicated deep residual network, and with accurate quantification of functional relative pressures achieved by subsequent physics-informed image processing. To achieve this, our two-step approach was trained and validated in a patient-specific in-silico cohort, showing good accuracy in estimating velocity (relative error: 15.0 ± 0.1%, mean absolute error (MAE): 0.07 ± 0.06 m/s, and cosine similarity: 0.99 ± 0.06 at peak velocity) and flow (relative error: 6.6 ± 4.7%, root mean square error (RMSE): 0.56 mL/s at peak flow), and with the coupled physics-informed image analysis allowing for maintained recovery of functional relative pressure throughout the circle of Willis (relative error: 11.0 ± 7.3%, RMSE: 0.3 ± 0.2 mmHg). Furthermore, the quantitative super-resolution approach is applied to an in-vivo volunteer cohort, effectively generating intracranial flow images at <0.5 mm resolution and showing reduced low-resolution bias in relative pressure estimation. Our work thus presents a promising two-step approach to non-invasively quantify cerebrovascular hemodynamics, being applicable to dedicated clinical cohorts in the future.",

keywords = "Cerebrovasculature, Deep learning, Relative pressure, Super-resolution 4D flow MRI",

author = "E. Ferdian and D. Marlevi and J. Schollenberger and M. Aristova and Edelman, {E. R.} and S. Schnell and Figueroa, {C. A.} and Nordsletten, {D. A.} and Young, {A. A.}",

note = "Publisher Copyright: {\textcopyright} 2023",

year = "2023",

month = aug,

doi = "10.1016/j.media.2023.102831",

language = "English (US)",

volume = "88",

journal = "Medical Image Analysis",

issn = "1361-8415",

publisher = "Elsevier B.V.",

}

Ferdian, E, Marlevi, D, Schollenberger, J, Aristova, M, Edelman, ER, Schnell, S, Figueroa, CA, Nordsletten, DA & Young, AA 2023, 'Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure', Medical Image Analysis, vol. 88, 102831. https://doi.org/10.1016/j.media.2023.102831

Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure. / Ferdian, E.; Marlevi, D.; Schollenberger, J. et al.
In: Medical Image Analysis, Vol. 88, 102831, 08.2023.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure

AU - Ferdian, E.

AU - Marlevi, D.

AU - Schollenberger, J.

AU - Aristova, M.

AU - Edelman, E. R.

AU - Schnell, S.

AU - Figueroa, C. A.

AU - Nordsletten, D. A.

AU - Young, A. A.

PY - 2023/8

Y1 - 2023/8

N2 - The development of cerebrovascular disease is tightly coupled to regional changes in intracranial flow and relative pressure. Image-based assessment using phase contrast magnetic resonance imaging has particular promise for non-invasive full-field mapping of cerebrovascular hemodynamics. However, estimations are complicated by the narrow and tortuous intracranial vasculature, with accurate image-based quantification directly dependent on sufficient spatial resolution. Further, extended scan times are required for high-resolution acquisitions, and most clinical acquisitions are performed at comparably low resolution (>1 mm) where biases have been observed with regard to the quantification of both flow and relative pressure. The aim of our study was to develop an approach for quantitative intracranial super-resolution 4D Flow MRI, with effective resolution enhancement achieved by a dedicated deep residual network, and with accurate quantification of functional relative pressures achieved by subsequent physics-informed image processing. To achieve this, our two-step approach was trained and validated in a patient-specific in-silico cohort, showing good accuracy in estimating velocity (relative error: 15.0 ± 0.1%, mean absolute error (MAE): 0.07 ± 0.06 m/s, and cosine similarity: 0.99 ± 0.06 at peak velocity) and flow (relative error: 6.6 ± 4.7%, root mean square error (RMSE): 0.56 mL/s at peak flow), and with the coupled physics-informed image analysis allowing for maintained recovery of functional relative pressure throughout the circle of Willis (relative error: 11.0 ± 7.3%, RMSE: 0.3 ± 0.2 mmHg). Furthermore, the quantitative super-resolution approach is applied to an in-vivo volunteer cohort, effectively generating intracranial flow images at <0.5 mm resolution and showing reduced low-resolution bias in relative pressure estimation. Our work thus presents a promising two-step approach to non-invasively quantify cerebrovascular hemodynamics, being applicable to dedicated clinical cohorts in the future.

AB - The development of cerebrovascular disease is tightly coupled to regional changes in intracranial flow and relative pressure. Image-based assessment using phase contrast magnetic resonance imaging has particular promise for non-invasive full-field mapping of cerebrovascular hemodynamics. However, estimations are complicated by the narrow and tortuous intracranial vasculature, with accurate image-based quantification directly dependent on sufficient spatial resolution. Further, extended scan times are required for high-resolution acquisitions, and most clinical acquisitions are performed at comparably low resolution (>1 mm) where biases have been observed with regard to the quantification of both flow and relative pressure. The aim of our study was to develop an approach for quantitative intracranial super-resolution 4D Flow MRI, with effective resolution enhancement achieved by a dedicated deep residual network, and with accurate quantification of functional relative pressures achieved by subsequent physics-informed image processing. To achieve this, our two-step approach was trained and validated in a patient-specific in-silico cohort, showing good accuracy in estimating velocity (relative error: 15.0 ± 0.1%, mean absolute error (MAE): 0.07 ± 0.06 m/s, and cosine similarity: 0.99 ± 0.06 at peak velocity) and flow (relative error: 6.6 ± 4.7%, root mean square error (RMSE): 0.56 mL/s at peak flow), and with the coupled physics-informed image analysis allowing for maintained recovery of functional relative pressure throughout the circle of Willis (relative error: 11.0 ± 7.3%, RMSE: 0.3 ± 0.2 mmHg). Furthermore, the quantitative super-resolution approach is applied to an in-vivo volunteer cohort, effectively generating intracranial flow images at <0.5 mm resolution and showing reduced low-resolution bias in relative pressure estimation. Our work thus presents a promising two-step approach to non-invasively quantify cerebrovascular hemodynamics, being applicable to dedicated clinical cohorts in the future.

KW - Cerebrovasculature

KW - Deep learning

KW - Relative pressure

KW - Super-resolution 4D flow MRI

UR - http://www.scopus.com/inward/record.url?scp=85160206323&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85160206323&partnerID=8YFLogxK

U2 - 10.1016/j.media.2023.102831

DO - 10.1016/j.media.2023.102831

M3 - Article

C2 - 37244143

AN - SCOPUS:85160206323

SN - 1361-8415

VL - 88

JO - Medical Image Analysis

JF - Medical Image Analysis

M1 - 102831

ER -

Ferdian E, Marlevi D, Schollenberger J, Aristova M, Edelman ER, Schnell S et al. Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure. Medical Image Analysis. 2023 Aug;88:102831. doi: 10.1016/j.media.2023.102831

Cerebrovascular super-resolution 4D Flow MRI – Sequential combination of resolution enhancement by deep learning and physics-informed image processing to non-invasively quantify intracranial velocity, flow, and relative pressure

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this