Assessment of the lung parenchyma by means of magnetic resonance images
A method for the assessment of the lung parenchyma in a human or an animal is indicated using a series of magnetic resonance (MR) images of the lung parenchyma acquired at different breathing positions in the same human or animal. The method comprises at least the steps of a.) estimating a change of the lung volume V L between the different breathing positions, b.) determining a signal intensity SI({right arrow over (x)} i ) in at least one same region or position {right arrow over (x)} i of the lung parenchyma for each of the MR images, and c.) determining at least one respiratory index α({right arrow over (x)} i ) according to the formula α ( x → i ) = - d ( log ( SI ( x → i ) ) ) d ( log ( V L ) ) . ( Fig . 2 )
1. A method for the assessment of the lung parenchyma in a human or an animal using a series of magnetic resonance (MR) images of the lung parenchyma acquired at different breathing positions in the same human or animal, comprising the following steps:
a.) estimating a change of a lung volume V L between the different breathing positions;
b.) determining a signal intensity SI({right arrow over (x)} i ) in at least one region or position {right arrow over (x)} i of the lung parenchyma for each image of the series of MR images, the at least one region or position {right arrow over (x)} i of the lung parenchyma being the same for each image of the series of MR images;
c.) determining at least one respiratory index α({right arrow over (x)} i ) according to the formula
α
(
x
→
i
)
=
-
d
(
log
(
SI
(
x
→
i
)
)
)
d
(
log
(
V
L
)
)
;
and
d.) assessing the function of the lung parenchyma at a plurality of respective regions or positions {right arrow over (x)} i of the lung based on the determined at least one respiratory index α({right arrow over (x)} i ).
2. The method as claimed in claim 1 ,
wherein the images of the acquired series of MR images are spatially co-registered in the area of the lung parenchyma, and
wherein, in step b.), the signal intensity SI({right arrow over (x)} i ) is determined for each of the spatially co-registered images of the series of MR images.
3. The method as claimed in claim 2 , wherein the spatial co-registration of the images of the series of MR images is carried out by means of a deformable B-spline mass preserving image registration algorithm.
4. The method as claimed in claim 2 ,
wherein the spatially co-registered images of the series of MR images are median-filtered, and wherein, in step b.), the signal intensity SI({right arrow over (x)} i ) is determined for each of the spatially co-registered and median-filtered images of the series of MR images.
5. The method as claimed in claim 1 ,
wherein the estimation of the lung volume V L for each of the different breathing positions is based on the acquired series of MR images.
6. The method as claimed in claim 5 ,
wherein each image of the acquired series of MR images is segmented with respect to the lung parenchyma, and wherein, in step a.), the lung volume V L is estimated for each of the different breathing positions based on the segmented images of the series of MR images.
7. The method as claimed in claim 6 ,
wherein the images of the acquired series of MR images are median-filtered prior to the segmentation.
8. The method as claimed in claim 1 ,
wherein the series of MR images comprises at least three images, and wherein, in step c.), the at least one respiratory index α({right arrow over (x)} i ) is determined by means of a linear fit.
9. The method as claimed in claim 1 ,
wherein steps b.) and c.) are carried out for a plurality of regions or positions {right arrow over (x)} i of the lung parenchyma.
10. The method as claimed in claim 9 ,
wherein the determined respiratory indices α({right arrow over (x)} i ) are used to generate a two-dimensional or three-dimensional map.
11. The method as claimed in claim 1 ,
wherein, in the determination of the at least one respiratory index α({right arrow over (x)} i ) in step c.), a multiplication with the factor
1
1
-
SNR
-
1
is miner carried out, in order to yield the formula
α
(
x
→
i
)
=
-
1
1
-
SNR
-
1
·
d
(
log
(
SI
(
x
→
i
)
)
)
d
(
log
(
V
L
)
)
,
wherein SNR denotes an estimated signal-to-noise ratio SI({right arrow over (x)} i )/η in which η represents the estimated intensity of noise in the series of MR images in the at least one region or position {right arrow over (x)} i .
12. The method as claimed in claim 1 , additionally comprising, before step b.), the step that
a series of MR images of the lung parenchyma is acquired at different breathing positions by means of a MR imaging system.
13. The method as claimed in claim 12 ,
wherein the series of MR images is acquired by means of a balanced steady state free precession (SSFP) imaging sequence.
14. A computer program product for the assessment of the lung parenchyma in a human or an animal, the computer program product comprising a non-transitory computer-readable medium comprising executable instructions to cause a processor of a computer to carry out the following steps by means of a series of magnetic resonance (MR) images of the lung parenchyma acquired at different breathing positions in the same human or animal:
i.) determining a signal intensity SI({right arrow over (x)} i ) in at least one region or position {right arrow over (x)} i of the lung parenchyma for each image of the series of MR images, the at least one region or position {right arrow over (x)} i of the lung parenchyma being the same for each image of the series of MR images; and
ii.) determining at least one respiratory index α({right arrow over (x)} i ) according to the formula
α
(
x
→
i
)
=
-
d
(
log
(
SI
(
x
→
i
)
)
)
d
(
log
(
V
L
)
)
,
wherein V L denotes a lung volume for each of the different breathing positions.
15. The computer program product as claimed in claim 14 , additionally comprising executable instructions to:
acquire, before step i.), the series of MR images of the lung parenchyma at different breathing positions by means of a MR imaging system and to
estimate, before step ii.), a change of the lung volume V L between the different breathing positions based on the acquired series of MR images.
16. The computer program product as claimed in claim 14 ,
wherein, in the determination of the at least one respiratory index α({right arrow over (x)} i ) in step ii.), a multiplication with the factor
1
1
-
SNR
-
1
is further carried out, in order to yield the formula
α
(
x
→
i
)
=
-
1
1
-
SNR
-
1
·
d
(
log
(
SI
(
x
→
i
)
)
)
d
(
log
(
V
L
)
)
,
wherein SNR denotes an estimated signal-to-noise ratio SI({right arrow over (x)} i )/η in which η represents the estimated intensity of noise in the series of MR images in the at least one region or position {right arrow over (x)} i .
17. A magnetic resonance imaging (MRI) system comprising:
a main magnet for generating a main magnetic field at a location of a sample to be imaged, in order to at least partly align nuclear spins of the sample;
an excitation module for applying radio frequency (RF) pulses to the sample, in order to repeatedly excite the nuclear spins of the sample;
a gradient module for generating temporary magnetic gradient fields at a location of the sample;
an acquisition module for acquiring the magnetic resonance signals produced by excited nuclear spins of the sample;
a control module configured for controlling the excitation module, the gradient module and the acquisition module such, that a series of MR images of the lung parenchyma of a human or an animal can be acquired at different breathing positions; and
an analysis module being configured for determining a signal intensity SI({right arrow over (x)} i ) in at least one region or position {right arrow over (x)} i of the lung parenchyma for each image of the series of MR images, the at least one region or position {right arrow over (x)} i of the lung parenchyma being the same for each image of the series of MR images, and being configured for determining at least one respiratory index α({right arrow over (x)} i ) according to the formula
α
(
x
→
i
)
=
-
d
(
log
(
SI
(
x
→
i
)
)
)
d
(
log
(
V
L
)
)
,
wherein V L denotes a lung volume for each of the different breathing positions.
18. The MRI system as claimed in claim 17 , the analysis module further being configured for estimating a change of the lung volume V L between the different breathing positions based on the series of MR images.
19. The MRI system as claimed in claim 17 ,
wherein the analysis module is further configured for carrying out a multiplication with the factor
1
1
-
SNR
-
1
in the determination of the at least one respiratory index α({right arrow over (x)} i ), in order to yield the formula
α
(
x
→
i
)
=
-
1
1
-
SNR
-
1
·
d
(
log
(
SI
(
x
→
i
)
)
)
d
(
log
(
V
L
)
)
,
wherein SNR denotes an estimated signal-to-noise ratio SI({right arrow over (x)} i )/η in which η represents the estimated intensity of noise in the series of MR images in the at least one region or position {right arrow over (x)} i .