Diffusion-weighted magnetic resonance imaging
A method of performing diffusion-weighted magnetic resonance imaging is provided. The method includes a step of: using a neural network to filter a diffusion-weighted image of an object acquired by a magnetic resonance imaging scanner, the neural network being programmed to produce an output image from the acquired image. The neural network improves the signal to noise ratio of the output image relative to the acquired image. The neural network, when applied to a synthetic knife-edge image to which Rician noise providing a signal-to-noise ratio of 13 or more is added, forms a curve of normalised values of modulation-transfer-function against frequency which has a higher area thereunder than the area under the corresponding curve of normalised values of modulation-transfer-function against frequency for a reference Gaussian smoothing filter.
1. A method of performing diffusion-weighted magnetic resonance imaging, the method including a step of:
using a neural network to filter a diffusion-weighted image of an object acquired by a magnetic resonance imaging scanner, the neural network being programmed to produce an output image from the acquired image;
wherein the neural network improves a signal to noise ratio of the output image relative to the acquired image; and
wherein the neural network, when applied to a synthetic knife-edge image to which Rician noise providing a signal-to-noise ratio of 13 or more is added, forms a curve of normalised values of modulation-transfer-function against frequency which has a higher area thereunder than an area under a corresponding curve of normalised values of modulation-transfer-function against frequency for a reference Gaussian smoothing filter, {tilde over (G)}:
G
~
{
f
(
x
,
y
)
}
=
1
2
π
σ
2
exp
{
-
(
x
2
+
y
2
)
2
σ
2
}
f
(
x
,
y
)
→
g
(
x
,
y
)
where f(x, y) is the noisy synthetic knife-edge image, g(x, y) is the filtered diffusion-weighted image, is a convolution parameter, and σ 2 is a smoothing variance set such that σ 2 =4.
2. The method according to claim 1 wherein the acquired image is acquired by the magnetic resonance imaging scanner at a given b-value and a given orientation for that b-value and without signal averaging.
3. The method according to claim 1 wherein the neural network is a spatially variant filter.
4. The method according to claim 1 wherein the neural network is a convolutional neural network.
5. The method according to claim 1 , including a preliminary step of:
acquiring the diffusion-weighted image using the magnetic resonance imaging scanner.
6. The method according to claim 1 wherein the acquired image is acquired at a first b-value and at given location in the object, and the method includes a further step of:
combining the output image with one or more further diffusion-weighted images acquired at the given location in the object but at respective different b-values to derive an image of an apparent diffusion coefficient of water at the given location in the object.
7. The method according to claim 1 wherein the object is a human or animal subject.
8. The method according to claim 7 , including a further step of:
analysing the output image or an image derived therefrom for assessment of disease extent in the human or animal subject.
9. A computer program, stored on a non-transitory computer readable medium, that when executed, causes a computer to perform the method of claim 1 .
10. An imaging system for performing diffusion-weighted magnetic resonance imaging, the system including:
a magnetic resonance imaging scanner for acquiring a diffusion-weighted image of an object; and
a computer system which receives the acquired image, and is programmed with a neural network which filters the acquired image to produce an output image from the acquired image;
wherein the neural network improves a signal to noise ratio of the output image relative to the acquired image by a factor of X as measured by Y; and
wherein the neural network, when applied to a synthetic knife-edge image to which Rician noise providing a signal-to-noise ratio of 13 or more is added, forms a curve of normalised values of modulation-transfer-function against frequency which has a higher area thereunder than an area under a corresponding curve of normalised values of modulation-transfer-function against frequency for a reference Gaussian smoothing filter, {tilde over (G)}:
G
~
{
f
(
x
,
y
)
}
=
1
2
π
σ
2
exp
{
-
(
x
2
+
y
2
)
2
σ
2
}
f
(
x
,
y
)
→
g
(
x
,
y
)
where f(x, y) is the noisy synthetic knife-edge image, g(x, y) is the filtered acquired image, is a convolution parameter, and σ 2 is a smoothing variance set such that σ 2 =4.
11. A computer system which receives a diffusion-weighted image of an object acquired by a magnetic resonance imaging scanner, and is programmed with a neural network which filters the acquired image to produce an output image from the acquired image;
wherein the neural network improves a signal to noise ratio of the output image relative to the acquired image by a factor of X as measured by Y; and
wherein the neural network, when applied to a synthetic knife-edge image to which Rician noise providing a signal-to-noise ratio of 13 or more is added, forms a curve of normalised values of modulation-transfer-function against frequency which has a higher area thereunder than an area under a corresponding curve of normalised values of modulation-transfer-function against frequency for a reference Gaussian smoothing filter, {tilde over (G)}:
G
~
{
f
(
x
,
y
)
}
=
1
2
π
σ
2
exp
{
-
(
x
2
+
y
2
)
2
σ
2
}
f
(
x
,
y
)
→
g
(
x
,
y
)
where f(x, y) is the noisy synthetic knife-edge image, g(x, y) is the filtered acquired image, is a convolution parameter, and σ 2 is a smoothing variance set such that σ 2 =4.