TY - JOUR

T1 - Noise estimation for the velocity in MRI phase-contrast

AU - Irarrazaval, Pablo

AU - Dehghan Firoozabadi, Ali

AU - Uribe, Sergio

AU - Tejos, Cristian

AU - Sing-Long, Carlos

N1 - Funding Information:
The authors acknowledge financial support from: FONDECYT N o 3190147 , CONICYT -PIA-Anillo ACT1416 , FONDECYT N o 1191710 , and FONDECYT N o 1181057 . Also, Carlos Sing-Long was partially funded by FONDECYT N o 11160728 . The Millennium Nucleus for Cardiovascular Magnetic Resonance is supported by the Millennium Scientific Initiative of the Ministry of Economy, Development and Tourism (Chile).
Funding Information:
The authors acknowledge financial support from: FONDECYT No 3190147, CONICYT-PIA-Anillo ACT1416, FONDECYT No 1191710, and FONDECYT No 1181057. Also, Carlos Sing-Long was partially funded by FONDECYT No 11160728. The Millennium Nucleus for Cardiovascular Magnetic Resonance is supported by the Millennium Scientific Initiative of the Ministry of Economy, Development and Tourism (Chile).
Publisher Copyright:
© 2019 Elsevier Inc.

PY - 2019/11

Y1 - 2019/11

N2 - The purpose of this study is to estimate the precision or statistical variability of the velocity measurements computed from MRI phase-contrast. From the analytical probability density function (PDF) of the phase in the signal we obtain the PDF of the velocity by means of an auto-convolution. This PDF allows the estimation of the precision of the velocity, important for the correct interpretation of the many parameters that are based on it. We show that for high Signal-to-Noise Ratio (SNR) voxels, the distribution is well approximated by a Gaussian distribution. On the other hand, this is not true for lower SNR voxels, where the distribution adopts a form in between the Gaussian and the uniform distributions. This was confirmed empirically. Also, knowing the PDF on a coil by coil basis it is possible to combine the data from multiple coils in an optimal way. We showed that the optimal combination reduces the resulting global variability of the velocity, in comparison with the commonly used Weighted Mean or with a SENSE reconstruction with R = 1.

AB - The purpose of this study is to estimate the precision or statistical variability of the velocity measurements computed from MRI phase-contrast. From the analytical probability density function (PDF) of the phase in the signal we obtain the PDF of the velocity by means of an auto-convolution. This PDF allows the estimation of the precision of the velocity, important for the correct interpretation of the many parameters that are based on it. We show that for high Signal-to-Noise Ratio (SNR) voxels, the distribution is well approximated by a Gaussian distribution. On the other hand, this is not true for lower SNR voxels, where the distribution adopts a form in between the Gaussian and the uniform distributions. This was confirmed empirically. Also, knowing the PDF on a coil by coil basis it is possible to combine the data from multiple coils in an optimal way. We showed that the optimal combination reduces the resulting global variability of the velocity, in comparison with the commonly used Weighted Mean or with a SENSE reconstruction with R = 1.

KW - Flow MRI

KW - Phase contrast velocity

KW - Velocity-to-Noise Ratio

UR - http://www.scopus.com/inward/record.url?scp=85071398193&partnerID=8YFLogxK

U2 - 10.1016/j.mri.2019.08.028

DO - 10.1016/j.mri.2019.08.028

M3 - Article

C2 - 31449850

AN - SCOPUS:85071398193

SN - 0730-725X

VL - 63

SP - 250

EP - 257

JO - Magnetic Resonance Imaging

JF - Magnetic Resonance Imaging

ER -