Super-resolution photoacoustic microscopy
A method for super-resolution photoacoustic microscopy of an object. The method includes optically exciting the object according to a plurality of excitation patterns utilizing a digital micromirror device (DMD), receiving a plurality of acoustic waves propagated from the object due to optically exciting the object, reconstructing each of a plurality of photoacoustic (PA) images from a respective acoustic wave of the plurality of acoustic waves, and obtaining a super-resolution PA image of the object from the plurality of PA images by applying a frequency domain reconstruction method to the plurality of PA images. Each of the plurality of acoustic waves are associated with a respective excitation pattern of the plurality of excitation patterns.
1. A method for super-resolution photoacoustic microscopy of an object, the method comprising:
emitting a laser beam to a digital micromirror device (DMD) utilizing a laser;
generating a plurality of modulated light beams from the laser beam by producing each of a plurality of periodic square patterns (PSPs) at a respective orientation of a plurality of orientations and a respective phase shift of a plurality of phase shifts via the DMD;
generating a plurality of sinusoidal spatial fringe (SSF) patterns from the plurality of PSPs by filtering the plurality of modulated light beams utilizing an optical filter;
optically exciting the object by:
emitting each of the plurality of SSF patterns to the object; and
reflecting the laser beam to the object by the DMD;
receiving, utilizing an ultrasound transducer (UST), a plurality of acoustic waves propagated from the object due to optically exciting the object, each of the plurality of acoustic waves associated with a respective orientation of a plurality of orientations and a respective phase shift of a plurality of phase shifts;
reconstructing, utilizing one or more processors, each of a plurality of photoacoustic (PA) images from a respective acoustic wave of the plurality of acoustic waves;
obtaining, utilizing the one or more processors, a plurality of filtered PA images by passing each of the plurality of PA images through a low-pass filter of a first cut-off frequency equal to a spatial spectral bandwidth of the UST;
obtaining, utilizing the one or more processors, a plurality of down-shifted components from the plurality of filtered PA images by:
calculating a first down-shifted component {circumflex over (M)} a (k x −k x 1 ,k y −k y 1 ,z) of the plurality of down-shifted components at an orientation θ of the plurality of orientations according to an operation defined by the following:
M
^
a
(
k
x
-
k
x
1
,
k
y
-
k
y
1
,
z
)
=
π
4
η
Γ
[
(
P
^
0
0
(
k
x
,
k
y
,
z
)
-
P
^
0
π
(
k
x
,
k
y
,
z
)
)
+
i
(
[
1
+
8
π
2
]
P
^
0
NI
(
k
x
,
k
y
,
z
)
-
P
^
0
0
(
k
x
,
k
y
,
z
)
-
2
P
^
0
π
/
2
(
k
x
,
k
y
,
z
)
-
P
^
0
π
(
k
x
,
k
y
,
z
)
)
]
where:
{circumflex over (P)} 0 NI (k x ,k y ,z) is a representation of a respective PA image of the plurality of PA images in a wavenumber domain (k x ,k y ) that is reconstructed, utilizing the UST, from a respective acoustic wave of the plurality of acoustic waves that is generated at a depth z of the object due to reflecting the laser beam to the object by the DMD,
{circumflex over (P)} 0 φ 0 (k x ,k y ,z) is a representation of a respective PA image of the plurality of PA images in the wavenumber domain (k x ,k y ) where the respective PA image is reconstructed, utilizing the UST, from a respective acoustic wave of the plurality of acoustic waves that is generated at the depth z of the object due to emitting a respective SSF pattern of the plurality of SSF patterns to the object, wherein the respective SSF pattern is generated from a respective PSP of the plurality of PSPs that is produced at the orientation θ and a phase shift φ 0 of the plurality of phase shifts,
η is a heat conversion efficiency of the object,
Γ is the Gruneisen parameter,
i is the imaginary unit, and
k x 1 k y 1 are spatial frequencies of the respective SSF pattern given by:
k x 1 =2π f SSF cos(θ), and
k y 1 = 2 πf SSF sin(θ),
where f SSF is a frequency of the plurality of SSF patterns; and
calculating a second down-shifted component of the plurality of down-shifted components at an orientation π+θ of the plurality of orientations by flipping and conjugating of the first down-shifted component;
extracting, utilizing the one or more processors, a plurality of high-frequency components from the plurality of down-shifted components by passing the plurality of down-shifted components through a high-pass filter of a second cut-off frequency equal to the spatial spectral bandwidth of the UST; and
generating, utilizing the one or more processors, a super-resolution PA image of the object by applying an inverse Fourier transform to the plurality of high-frequency components.
2. A method for super-resolution photoacoustic microscopy of an object, the method comprising:
optically exciting the object according to a plurality of excitation patterns utilizing a digital micromirror device (DMD), each of the plurality of excitation patterns associated with a respective orientation of a plurality of orientations and a respective phase shift of a plurality of phase shifts;
receiving, utilizing an ultrasound transducer (UST), a plurality of acoustic waves propagated from the object due to optically exciting the object, each of the plurality of acoustic waves associated with a respective excitation pattern of the plurality of excitation patterns; and
reconstructing, utilizing one or more processors, each of a plurality of photoacoustic (PA) images from a respective acoustic wave of the plurality of acoustic waves; and
obtaining, utilizing the one or more processors, a super-resolution PA image of the object from the plurality of PA images by applying a frequency domain reconstruction method to the plurality of PA images.
3. The method of claim 2 , wherein optically exciting the object comprises:
emitting a laser beam to the DMD utilizing a laser;
generating a plurality of modulated light beams from the laser beam by producing a plurality of periodic square patterns (PSPs) via the DMD, each of the plurality of PSPs comprising a respective orientation of the plurality of orientations and a respective phase shift of the plurality of phase shifts;
generating a plurality of sinusoidal spatial fringe (SSF) patterns from the plurality of PSPs by filtering the plurality of modulated light beams utilizing an optical filter; and
exciting the object by emitting each of the plurality of SSF patterns to the object.
4. The method of claim 3 , wherein producing the plurality of PSPs comprises producing each of the plurality of PSPs at a respective orientation of the plurality of orientations and a respective phase shift of the plurality of phase shifts.
5. The method of claim 4 , wherein optically exciting the object further comprises reflecting the laser beam to the object by the DMD.
6. The method of claim 5 , wherein applying the frequency domain reconstruction method comprises:
obtaining a plurality of filtered PA images by low-pass filtering each of the plurality of PA images;
obtaining a plurality of down-shifted components from the plurality of filtered PA images by applying a phase compounding method to the plurality of filtered PA images, each of the plurality of down-shifted components associated with a respective orientation of the plurality of orientations;
extracting a plurality of high-frequency components from the plurality of down-shifted components by high-pass filtering the plurality of down-shifted components; and
generating the super-resolution PA image by applying an inverse Fourier transform to the plurality of high-frequency components.
7. The method of claim 6 , wherein low-pass filtering each of the plurality of PA images comprises passing each of the plurality of PA images through a low-pass filter of a cut-off frequency equal to a spatial spectral bandwidth of the UST.
8. The method of claim 6 , wherein applying the phase compounding method to the plurality of filtered PA images comprises:
calculating a first down-shifted component {circumflex over (M)} α (k x −k x 1 ,k y −k y 1 ,z) of the plurality of down-shifted components at an orientation θ of the plurality of orientations according to an operation defined by the following:
M
^
a
(
k
x
-
k
x
1
,
k
y
-
k
y
1
,
z
)
=
π
4
η
Γ
[
(
P
^
0
0
(
k
x
,
k
y
,
z
)
-
P
^
0
π
(
k
x
,
k
y
,
z
)
)
+
i
(
[
1
+
8
π
2
]
P
^
0
NI
(
k
x
,
k
y
,
z
)
-
P
^
0
0
(
k
x
,
k
y
,
z
)
-
2
P
^
0
π
/
2
(
k
x
,
k
y
,
z
)
-
P
^
0
π
(
k
x
,
k
y
,
z
)
)
]
where:
{circumflex over (P)} 0 NI (k x ,k y ,z) is a representation of a respective PA image of the plurality of PA images in a wavenumber domain (k x ,k y ) that is reconstructed, utilizing the UST, from a respective acoustic wave of the plurality of acoustic waves that is generated at a depth z of the object due to reflecting the laser beam to the object by the DMD,
{circumflex over (P)} 0 φ 0 (k x ,k y ,z) is a representation of a respective PA image of the plurality of PA images in the wavenumber domain (k x ,k y ) where the respective PA image is reconstructed, utilizing the UST, from a respective acoustic wave of the plurality of acoustic waves that is generated at the depth z of the object due to emitting a respective SSF pattern of the plurality of SSF patterns to the object, wherein the respective SSF pattern is generated from a respective PSP of the plurality of PSPs that is produced at the orientation θ and a phase shift φ 0 of the plurality of phase shifts,
η is a heat conversion efficiency of the object,
Γ is the Gruneisen parameter,
i is the imaginary unit, and
k x 1 k y 1 are spatial frequencies of the respective SSF pattern given by:
k x 1 =2π f SSF cos(θ), and
k y 1 =2π f SSF sin(θ),
where f SSF is a frequency of the plurality of SSF patterns; and
calculating a second down-shifted component of the plurality of down-shifted components at an orientation π+θ of the plurality of orientations by flipping and conjugating of the first down-shifted component.
9. The method of claim 6 , wherein high-pass filtering the plurality of down-shifted components comprises passing the plurality of down-shifted components through a high-pass filter of a cut-off frequency equal to a spatial spectral bandwidth of the UST.
10. The method of claim 4 , wherein producing each of the plurality of PSPs at a respective orientation of the plurality of orientations comprises setting each of the plurality of orientations to one of 0, π/4, π/2, and 3π/4.
11. The method of claim 4 , wherein producing each of the plurality of PSPs at a respective phase shift of the plurality of phase shifts comprises shifting each of the plurality of PSPs on the DMD by a phase shift equal to one of 0, π/2, and π.
12. The method of claim 3 , wherein emitting the laser beam to the DMD comprises:
generating the laser beam by the laser;
reflecting the laser beam to a beam expander by a first mirror;
generating, utilizing the beam expander, an expanded laser beam from the laser beam by expanding a width of the laser beam; and
reflecting the expanded laser beam to the DMD by a second mirror.
13. The method of claim 12 , wherein expanding the width of the laser beam comprises:
converging, utilizing a first optical lens, the laser beam at a first focal point of the first optical lens; and
collimating, utilizing a second optical lens, a diverging light beam from the first focal point by placing a second focal point of the second optical lens at the first focal point.
14. The method of claim 3 , wherein filtering the plurality of modulated light beams comprises:
converging, utilizing a third optical lens, the plurality of modulated light beams at a third focal point of the third optical lens;
filtering high-order spatial frequencies of the plurality of modulated light beams by passing the plurality of modulated light beams through a pinhole placed at the third focal point; and
collimating, utilizing a fourth optical lens, a plurality of diverging light beams from the third focal point by placing a fourth focal point of the fourth optical lens at the third focal point.
15. The method of claim 14 , wherein passing the plurality of modulated light beams through the pinhole comprises setting a radius of the pinhole equal to a second order spatial frequency component of the plurality of SSF patterns according to an operation defined by the following:
r
p
=
λ
f
L
3
×
(
2
f
SSF
)
f
L
4
f
L
3
where:
r p is the radius of the pinhole,
λ is a wavelength of the plurality of SSF patterns,
f L3 is a focal length of the third optical lens,
f L4 is a focal length of the fourth optical lens, and
f SSF is a frequency of the plurality of SSF patterns given by f SSF =2πf L3 /f L4 .
16. The method of claim 14 , wherein collimating the plurality of diverging light beams comprises reflecting the plurality of diverging light beams to the fourth optical lens by a third mirror.