Determining position of magnetic resonance data with respect to magnetic field sensors
According to an example aspect of the present invention, there is provided generating, Low-Field-Magnetic Resonance Imaging, LF-MRI, or Ultra-Low-Field Magnetic Resonance Imaging, ULF-MRI, data with respect to an image frame, determining a sensorwise agreement of the data with determined sensitivity profiles, and determining a mapping between the image frame and a sensor frame, such that the sensorwise agreement has been fulfilled.
1. A method comprising:
generating, by a magnetic resonance imaging system comprising sensors arranged at positions around an imaged target volume, Low-Field-Magnetic Resonance Imaging, LF-MRI, or Ultra-Low-Field Magnetic Resonance Imaging, ULF-MRI, data with respect to an image frame;
determining, by the magnetic resonance imaging system, a sensorwise agreement of the data with determined sensitivity profiles; and
determining, by the magnetic resonance imaging system, a mapping between the image frame and a sensor frame, such that the sensorwise agreement has been fulfilled.
2. The method according to claim 1 , wherein the sensors are configured for magnetoencephalography, MEG, source localization, and the method further comprises:
obtaining, by the sensors one or more MEG data sets; and localizing at least one source of electrical brain activity of the MEG data sets based on the determined mapping.
3. The method according to claim 2 , wherein the sensor frame is maintained between generating the LF-MRI data or ULF-MRI data, and the MEG data.
4. The method according to claim 1 , wherein the mapping is determined iteratively using a nonlinear optimization method based on a subset of voxels.
5. The method according to claim 1 , wherein the mapping is determined based on solving a parametrized transfer function between voxel positions of the image frame and the sensor frame.
6. The method according to claim 1 , wherein the sensorwise agreement is based on an objective function for measuring similarity of LF-MRI data or ULF-MRI data and the sensitivity profiles of the sensors.
7. The method according to claim 6 , wherein the criterion comprises one or more or a combination of the following:
a target value of the objective function,
a gradient of the objective function satisfying at least one condition,
a change of the value of the objective function between iterations is sufficiently small,
a predetermined number of iterations for determining parameters for the mapping has been reached.
8. The method according to claim 6 , wherein the objective function is determined to reach a target value based on matching, or a similarity of, magnitudes and phases of the sensitivity profiles and the voxels.
9. The method according to claim 6 , wherein the agreement is determined based on an objective function, according to
g
(
p
)
=
∑
n
=
1
N
v
❘
"\[LeftBracketingBar]"
s
n
(
p
)
H
u
n
❘
"\[RightBracketingBar]"
s
(
p
)
u
,
where ∥⋅∥ denotes the Euclidean vector norm and (⋅) H the conjugate transpose, N v is a subset of voxels, s is a sensitivity vector, p is coordinate vector of parameters, u is a voxel vector comprising values of a subset of voxels and g is the objective function.
10. The method according to claim 1 , wherein the sensorwise agreement is fulfilled according to at least one criterion.
11. The method according to claim 1 , wherein the sensors comprise Superconducting QUantum Interference Devices, SQUIDs.
12. A magnetic resonance imaging system comprising sensors arranged at positions around an imaged target volume, comprising means for performing:
generating Low-Field-Magnetic Resonance Imaging, LF-MRI, or Ultra-Low-Field Magnetic Resonance Imaging, ULF-MRI, data with respect to an image frame;
determining a sensorwise agreement of the data with determined sensitivity profiles; and
determining a mapping between the image frame and a sensor frame, such that the sensorwise agreement has been fulfilled.
13. The magnetic resonance imaging system according to claim 12 , wherein the sensors are configured for magnetoencephalography, MEG, source localization, and the method comprises:
obtaining, by the sensors one or more MEG data sets; and localizing at least one source of electrical brain activity of the MEG data sets based on the determined mapping.
14. The magnetic resonance imaging system according to claim 13 , wherein the sensor frame is maintained between generating the LF-MRI data or ULF-MRI data, and the MEG data.
15. The magnetic resonance imaging system according to claim 12 , wherein the mapping is determined iteratively using a nonlinear optimization method based on a subset of voxels.
16. The magnetic resonance imaging system according to claim 12 , wherein the mapping is determined based on solving a parametrized transfer function between voxel positions of the image frame and the sensor frame.
17. The magnetic resonance imaging system according to claim 12 , wherein the sensorwise agreement is measured based on an objective function for measuring similarity of LF-MRI data or ULF-MRI data and the sensitivity profiles of the sensors.
18. The magnetic resonance imaging system according to claim 17 , wherein the criterion comprises one or more or a combination of the following:
a target value of the objective function,
a gradient of the objective function satisfying at least one condition,
a change of the value of the objective function between iterations is sufficiently small,
a predetermined number of iterations for determining parameters for the mapping has been reached.
19. The magnetic resonance imaging system according to claim 17 , wherein the target value of the objective function is determined to reach a target value based on matching, or a similarity of, magnitudes and phases of the sensitivity profiles and the voxels.
20. The magnetic resonance imaging system according to claim 17 , wherein the agreement is determined based on an objective function, according to
g
(
p
)
=
∑
n
=
1
N
v
❘
"\[LeftBracketingBar]"
s
n
(
p
)
H
u
n
❘
"\[RightBracketingBar]"
s
(
p
)
u
,
where ∥⋅∥ denotes the Euclidean vector norm and (⋅) H the conjugate transpose, N v is a subset of voxels, s is a sensitivity vector, p is coordinate vector of parameters, u is a voxel vector comprising values of a subset of voxels and g is the objective function.
21. The magnetic resonance imaging system according to claim 12 , wherein the sensorwise agreement is fulfilled according to at least one criterion.
22. The magnetic resonance imaging system according to claim 12 , wherein the sensors comprise Superconducting QUantum Interference Devices, SQUIDs.
23. The magnetic resonance imaging system according to claim 12 , wherein the magnetic resonance imaging system is a hybrid MEG-MRI device.
24. The magnetic resonance imaging system according to claim 12 , wherein the means comprises at least one processor; and at least one memory including computer program code, the at least one memory and computer program code configured to, with the at least one processor, cause the performance of the magnetic resonance imaging system.