Physics-Informed Deep Learning for Motion-Corrected Reconstruction of Quantitative Brain MRI

Hannah Eichhorn; Veronika Spieker; Kerstin Hammernik; Elisa Saks; Kilian Weiss; Christine Preibisch; Julia A. Schnabel

Physics-Informed Deep Learning for Motion-Corrected Reconstruction of Quantitative Brain MRI

Hannah Eichhorn, Veronika Spieker, Kerstin Hammernik, Elisa Saks, Kilian Weiss, Christine Preibisch, Julia A. Schnabel

Biomedical Computing

Research output: Working paper/Preprint › Preprint

6 Downloads (Pure)

Abstract

We propose PHIMO, a physics-informed learning-based motion correction method tailored to quantitative MRI. PHIMO leverages information from the signal evolution to exclude motion-corrupted k-space lines from a data-consistent reconstruction. We demonstrate the potential of PHIMO for the application of T2* quantification from gradient echo MRI, which is particularly sensitive to motion due to its sensitivity to magnetic field inhomogeneities. A state-of-the-art technique for motion correction requires redundant acquisition of the k-space center, prolonging the acquisition. We show that PHIMO can detect and exclude intra-scan motion events and, thus, correct for severe motion artifacts. PHIMO approaches the performance of the state-of-the-art motion correction method, while substantially reducing the acquisition time by over 40%, facilitating clinical applicability. Our code is available at https://github.com/HannahEichhorn/PHIMO.

Original language	Undefined/Unknown
Publication status	Published - 13 Mar 2024

Keywords

eess.IV

Access to Document

2403.08298v2

Cite this

@techreport{2f6bea4b831140e899461013825b0c3a,

title = "Physics-Informed Deep Learning for Motion-Corrected Reconstruction of Quantitative Brain MRI",

abstract = "We propose PHIMO, a physics-informed learning-based motion correction method tailored to quantitative MRI. PHIMO leverages information from the signal evolution to exclude motion-corrupted k-space lines from a data-consistent reconstruction. We demonstrate the potential of PHIMO for the application of T2* quantification from gradient echo MRI, which is particularly sensitive to motion due to its sensitivity to magnetic field inhomogeneities. A state-of-the-art technique for motion correction requires redundant acquisition of the k-space center, prolonging the acquisition. We show that PHIMO can detect and exclude intra-scan motion events and, thus, correct for severe motion artifacts. PHIMO approaches the performance of the state-of-the-art motion correction method, while substantially reducing the acquisition time by over 40%, facilitating clinical applicability. Our code is available at https://github.com/HannahEichhorn/PHIMO.",

keywords = "eess.IV",

author = "Hannah Eichhorn and Veronika Spieker and Kerstin Hammernik and Elisa Saks and Kilian Weiss and Christine Preibisch and Schnabel, {Julia A.}",

note = "Accepted at MICCAI 2024",

year = "2024",

month = mar,

day = "13",

language = "Undefined/Unknown",

type = "WorkingPaper",

}

TY - UNPB

T1 - Physics-Informed Deep Learning for Motion-Corrected Reconstruction of Quantitative Brain MRI

AU - Eichhorn, Hannah

AU - Spieker, Veronika

AU - Hammernik, Kerstin

AU - Saks, Elisa

AU - Weiss, Kilian

AU - Preibisch, Christine

AU - Schnabel, Julia A.

N1 - Accepted at MICCAI 2024

PY - 2024/3/13

Y1 - 2024/3/13

N2 - We propose PHIMO, a physics-informed learning-based motion correction method tailored to quantitative MRI. PHIMO leverages information from the signal evolution to exclude motion-corrupted k-space lines from a data-consistent reconstruction. We demonstrate the potential of PHIMO for the application of T2* quantification from gradient echo MRI, which is particularly sensitive to motion due to its sensitivity to magnetic field inhomogeneities. A state-of-the-art technique for motion correction requires redundant acquisition of the k-space center, prolonging the acquisition. We show that PHIMO can detect and exclude intra-scan motion events and, thus, correct for severe motion artifacts. PHIMO approaches the performance of the state-of-the-art motion correction method, while substantially reducing the acquisition time by over 40%, facilitating clinical applicability. Our code is available at https://github.com/HannahEichhorn/PHIMO.

AB - We propose PHIMO, a physics-informed learning-based motion correction method tailored to quantitative MRI. PHIMO leverages information from the signal evolution to exclude motion-corrupted k-space lines from a data-consistent reconstruction. We demonstrate the potential of PHIMO for the application of T2* quantification from gradient echo MRI, which is particularly sensitive to motion due to its sensitivity to magnetic field inhomogeneities. A state-of-the-art technique for motion correction requires redundant acquisition of the k-space center, prolonging the acquisition. We show that PHIMO can detect and exclude intra-scan motion events and, thus, correct for severe motion artifacts. PHIMO approaches the performance of the state-of-the-art motion correction method, while substantially reducing the acquisition time by over 40%, facilitating clinical applicability. Our code is available at https://github.com/HannahEichhorn/PHIMO.

KW - eess.IV

M3 - Preprint

BT - Physics-Informed Deep Learning for Motion-Corrected Reconstruction of Quantitative Brain MRI

ER -