In this work we present photoacoustic projection imaging with a 64-channel integrating line detector array, which average the pressure over cylindrical surfaces. For imaging, the line detectors are arranged parallel to each other on a cylindrical surface surrounding a specimen. Thereby, the three-dimensional imaging problem is reduced to a twodimensional problem, facilitating projection imaging. After acquisition of a dataset of pressure signals, a twodimensional photoacoustic projection image is reconstructed. The 64 channel line detector array is realized using optical fibers being part of interferometers. The parts of the interferometers used to detect the ultrasonic pressure waves consist of graded-index polymer-optical fibers (POFs), which exhibit better sensitivity than standard glass-optical fibers. Ultrasonic waves impinging on the POFs change the phase of light in the fiber-core due to the strain-optic effect. This phase shifts, representing the pressure signals, are demodulated using high-bandwidth balanced photo-detectors. The 64 detectors are optically multiplexed to 16 detection channels, thereby allowing fast imaging. Results are shown on a Rhodamine B dyed microsphere.
Most reconstruction algorithms for photoacoustic tomography, like back projection or time reversal, work ideally for point-like detectors. For real detectors, which integrate the pressure over their finite size, images reconstructed by these algorithms show some blurring. Iterative reconstruction algorithms using an imaging matrix can take the finite size of real detectors directly into account, but the numerical effort is significantly higher compared to the use of direct algorithms. For spherical or cylindrical detection surfaces, the blurring caused by a finite detector size is proportional to the distance from the rotation center (spin blur) and is equal to the detector size at the detection surface. In this work, we apply deconvolution algorithms to reduce this type of blurring on simulated and on experimental data. Two particular deconvolution methods are compared, which both utilize the fact that a representation of the blurred image in polar coordinates decouples pixels at different radii from the rotation center. Experimental data have been obtained with a flat, rectangular piezoelectric detector measuring signals around a plastisol cylinder containing various small photoacoustic sources with variable distance from the center. Both simulated and experimental results demonstrate a nearly complete elimination of spin blur.
Crack detection during continuous direct chill casting of aluminum is a matter of economics. Determining cracks during
production process saves money, energy and raw material. Of course, a non-destructive method is required for this
evaluation. Because of temperature concerns conventional ultrasound is not applicable. One non-contact alternative is
laser ultrasonics. In laser ultrasonics short laser pulses illuminate the sample. The electromagnetic energy gets absorbed
at the surface of the sample and results in local heating followed by expansion. Thereby broadband ultrasonic waves are
launched which propagate through the sample and get back reflected or scattered at interfaces (cracks, blowholes,…) like
conventional ultrasonic waves. Therefore laser ultrasonics is an alternative thermal infrared technology. By using an
interferometer also the detection of the ultrasonic waves at the sample surface is done in a remote manner. During
preliminary examinations in the lab by scanning different aluminum studs it was able to distinguish between studs with
and without cracks. The prediction of the dimension of the crack by evaluation of the damping of the broadband
ultrasonic waves was possible. With simple image reconstruction methods one can localize the crack and give an
estimation of its extent and even its shape. Subsequent first measurements using this laser ultrasonic setup during the
continuous casting of aluminum were carried out and showed the proof of principle in an industrial environment with
elevated temperatures, dust, cooling water and vibrations.
Detector arrays enable parallel detection for faster photoacoustic imaging than by moving a single detector, but the detector spacing for arrays cannot be smaller than the size of an array element. Spatial over-sampling is scanning with a step-size smaller than the size of the detector element and is possible only for a moving single detector. For a detector with finite sized surface the measured acoustic signal is a spatial average of the pressure field over the detector surface. If the reconstruction is performed assuming point-like detection over-sampling brings no advantage as e.g. for spherical or cylindrical detection surfaces the blurring caused by a finite detector size is proportional to the distance from the rotation center and is equal to the detector size at the detection surface.
Iterative reconstruction algorithms or inverting directly the imaging matrix can take the finite size of real detectors directly into account, but the numerical effort is significantly higher compared to direct algorithms assuming point-like detection. Another reconstruction with less numerical effort is to use a direct algorithm assuming point-like detectors and run a deconvolution algorithm for deblurring afterwards. For such reconstruction methods spatial over-sampling makes sense because it reduces the blurring significantly.
The effect of step size on the reconstructed image is systematically examined using simulated and experimental data. Experimental data are obtained on a plastisol cylinder with thin holes filled with an absorbing liquid. Data acquisition is done by utilization of a piezoelectric detector (PVDF stripe) which is rotated around the plastisol cylinder.
Most reconstruction algorithms for photoacoustic tomography, like back-projection or time-reversal, work ideally for point-like detectors. For real detectors, which integrate the pressure over their finite size, it was shown that images reconstructed by back-projection or time-reversal show some blurring. Iterative reconstruction algorithms using an imaging matrix can take the finite size of real detectors directly into account, but the numerical effort is significantly higher compared to the use of direct algorithms. For spherical or cylindrical detection surfaces the blurring caused by a finite detector size is proportional to the distance from the rotation center (“spin blur”) and is equal to the detector size at the detection surface. In this work we use deconvolution algorithms to reduce this type of blurring on simulated and on experimental data. Experimental data were obtained on a plastisol cylinder with 6 thin holes filled with an absorbing liquid (OrangeG). The holes were located on a spiral emanating from the center of the cylinder. Data acquisition was done by utilization of a piezoelectric detector which was rotated around the plastisol cylinder.
The recently introduced remote photoacoustic imaging technique allows measurement of photoacoustic signals on nonplanar
surfaces without the need for a water bath or coupling agent. Hereby, photoacoustically generated ultrasonic
displacements are detected without physical contact to the sample by utilizing laser interferometric techniques. In this
work we adapted different algorithms to allow reconstruction on non-planar surfaces and evaluate them on experimental
and simulated data. Experimental data were obtained using a remote photoacoustic setup based on two-wave mixing in a
photorefractive crystal. Ultrasonic displacements were acquired on flat and non-flat surfaces.
Three-dimensional reconstruction of simulated and real measurement data is shown with synthetic aperture focusing
technique, Fourier domain synthetic aperture focusing technique, and spectral-domain time reversal algorithms. For the
synthetic aperture focusing technique and the time reversal algorithm the surface morphology is taken into account. It is
demonstrated that artifacts can occur if the surface is not considered. For the experimental data the shape of the surface is
obtained from optical coherence tomography or by a priori knowledge.
We introduce a multichannel optical fiber based detector for photoacoustic imaging. By using in-house produced
photodetectors and relative low-cost components from telecommunication industries we were able to reduce the costs for
one channel significantly compared to previous setups. The estimated cost for one channel (without sampling device) is
below 800 €. The self-made balanced photodetector for 1550 nm achieves a gain of 100 dB, a -3dB bandwidth of 45
MHz and a maximum signal-to-noise-ratio of 48 dB. We present a four channel annular detector array based on optical
fiber Mach-Zehnder interferometers. Photoacoustic imaging is demonstrated by measuring photoacoustic signals of a
black polyethylene microsphere.
In this paper we report on remote photoacoustic imaging using an interferometric technique. By utilizing a two-wave
mixing interferometer ultrasonic displacements are measured without any physical contact to the sample.
This technique allows measurement of the displacements also on rough surfaces. Mixing a plane reference beam
with the speckled beam originating from the sample surface is done in a Bi12SiO20 photorefractive crystal.
After data acquisition the structure of the specimen is reconstructed using a Fourier-domain synthetic focusing
aperture technique. We show three-dimensional imaging on
tissue-mimicking phantoms and biological samples.
Furthermore, we show remote photoacoustic measurements on a human forearm in-vivo.
For photoacoustic imaging detectors which provide high spatial resolution while being highly sensitive are essential.
Integrating line detectors made of single mode polymer fibers achieve these requirements. In this paper
several approaches and preliminary experiments for single mode polymer fiber line detectors are presented. Operation
point stabilization by utilizing a fiber-based phase shifter is shown as well as results using different fiber
couplers in the setup.
As for any other imaging technique spatial resolution and sensitivity are important features for a photoacoustic imaging
device. It is already well known that spatial resolution depends on the size and the bandwidth of the detectors. Therefore
for photoacoustic image reconstruction usually small point-like and broadband detectors are assumed, which measure the
pressure as a function of time on a detection surface around the sample. But in reality point-like detectors are not ideal at
all: because of the small detector volume the thermodynamic fluctuations (= noise) get high and the signal amplitude is
low, which results in a bad signal-to-noise ratio (SNR). For a bigger detector volume the fluctuations are less and the
signal amplitude is better, which gives a better SNR. But on the other hand the photoacoustic pressure signal is averaged
over the whole detector volume, which results in blurring and a reduced spatial resolution if reconstruction algorithms
for point-like detectors are used. To characterize this trade-off between spatial resolution and sensitivity for a varying
detector volume in a quantitative way the pressure is described by a random variable having the measured pressure as a
mean value and noise as random fluctuations around that mean value ("stochastic process"). For a PVDF detector the
optimum for the detector size is given.
Photoacoustic imaging is an upcoming technique in the field of biomedical imaging. Our group introduced fiber-based
line detectors, which are used to acquire broad-band ultrasonic signals, several years ago. Up to now operating point
stabilization of fiber-based line detectors was realized by tuning the wavelength of the detection laser. This is, because
of the high costs, not applicable for parallel detection. An alternative stabilization method, the change of the optical path
length, is presented in this paper. Changing of the optical path length is realized by stretching the fiber with
piezoelectric tubes. Fringe patterns and operation point stabilization of both stabilization schemes are compared. Next,
signal detection utilizing a polymer optical fiber in a Mach-Zehnder and Fabry-Perot interferometer is demonstrated,
and the influence of the detection wavelength (633nm and 1550nm) is examined. Finally, two-dimensional imaging by
utilizing a perfluorinated polymer fiber is demonstrated.
Photoacoustic Imaging is an emerging imaging technology mainly for biology and medicine which combines
the advantages of diffuse optical imaging (high contrast) and ultrasonic imaging (high spatial resolution). A
short laser pulse excites the sample. The absorbed energy causes thermoelastic expansion and thereby launches
broadband ultrasonic waves. For collecting these waves we introduced integrating line detectors which integrate
the pressure along one direction. A fiber-based approach was realized using an interferometer integrated either
in glass fibers or polymer fibers. In this work we present the proof of principle of this approach.
Photoacoustic tomography is an emerging technology combining the advantages of optical imaging (high contrast)
and ultrasonic imaging (high spatial resolution). Applications for photoacoustic tomography are mainly
in imaging soft tissue. For photoacoustic imaging the sample is illuminated by a short pulse of electromagnetic
energy. Depending on the specific absorption rate (SAR) the electromagnetic radiation is absorbed and the
subsequent thermoelastic expansion launches broadband ultrasonic waves. Usually point like piezo-electric detectors
are used. Our group introduced integrating detectors a few years ago. This type of detector integrates
the pressure at least along one dimension. Integrating line detectors, which integrate the pressure along one
dimension, can be realized by using either free-beam or fiber-based interferometers. The latter approach also
allows other detector shapes than a line. In this paper we use a fiber-based annular detector for tomography.
Thereby the sample is rotated inside the annular detector on a position different from the symmetry axis of the
annular detector. Hence the sample is enclosed by the detector and all data from one plane are collected at
once. By moving the detector parallel to the symmetrie axis of the ring one can acquire data for a 3D image
reconstruction. Therfore, tomography can be performed with only one rotation axis and one translation axis.
For image reconstruction a novel algorithm is necessary which was tested on simulated data. Here we present an
imaging setup using such a fiber-based annular detector. First measurements of simple structures and subsequent
image reconstruction from these real data are shown in this paper.
Ultrasonic attenuation in biomaterials limits the quality and resolution of ultrasonic imaging. This work presents a simple
and reliable method to investigate acoustic attenuation of biological tissue samples and liquids in order to improve
reconstruction algorithms for photoacoustic imaging. For this purpose broadband high-frequency single transmission
measurements were performed. The spectra of the acquired signals were compared to reference measurements in
distilled water. Unfocused broadband piezoelectric transducers were used as ultrasound source and detector. Moreover,
laser generated ultrasound, which provides more intensity and signals with higher bandwidth, was used to measure
acoustic attenuation. Only few studies concerned with attenuation of fat tissue performed broadband high frequency
measurements and to our knowledge none of those used the simple and reliable single transmission approach with
unfocused ultrasound. Our results for acoustic attenuation in olive oil show good agreement with literature. Many studies
indicate linear frequency increase of attenuation of fat tissue. However, we observed significant non-linear frequency
behaviour of porcine subcutaneous fat tissue at room temperature with a power-law exponent of around 1.45.
For photoacoustic imaging, usually point-like detectors are used. As a special sensing technology for photoacoustic imaging, integrating detectors have been investigated that integrate the acoustic pressure over an area or line that is larger than the imaged object. Different kinds of optical fiber-based detectors are compared regarding their sensitivity and resolution in three-dimensional photoacoustic tomography. In the same type of interferometer, polymer optical fibers yielded much higher sensitivity than glass fibers. Fabry-Pérot glass-fiber interferometers in turn gave higher sensitivity than Mach-Zehnder-type interferometers. Regarding imaging resolution, the single-mode glass fiber showed the best performance. Last, three-dimensional images of phantoms and insects using a glass-fiber-based Fabry-Pérot interferometer as integrating line detector are presented.
Photoacoustic imaging is a novel imaging method for medical and biological applications, combining the advantages
of Diffuse Optical Imaging (high contrast) and Ultrasonic Imaging (high spatial resolution). A short laser
pulse hits the sample. The absorbed energy causes a thermoelastic expansion and thereby launches a broadband
ultrasonic wave (photoacoustic signal). The distribution of absorbed energy density is reconstructed from
measurements of the photoacoustic signals around the sample. For collecting photoacoustic signals either point
like or extended, integrating detectors can be used. The latter integrate the pressure at least in one dimension,
e.g. along a line. Thereby, the three dimensional imaging problem is reduced to a two dimensional problem.
For a tomography device consisting of a scanning line detector and a rotating sample, fiber-based detectors
made of polymer have been recently introduced. Fiber-based detectors are easy to use and possess a constant,
high spatial resolution over their entire active length. Polymer fibers provide a better impedance matching and
a better handling compared with glass fibers which were our first approach. First measurement results using
polymer fiber detectors and some approaches for improving the performance are presented.
In this work measurements of the acoustic attenuation coefficient for water in the frequency range of 20-40 Mhz have
been performed. These measurements have been conducted with a photoacoustic setup, containing a nano second pulse
laser and an integrating line detector. The nanosecond pulse laser is used to generate ultra sound waves in a target. Those
waves are transmitted into water and are then detected by the line sensor. In this way the resulting ultra sound field was
scanned and the absorption coefficient was evaluated. Different pulse energies, beam diameters and target materials were
used. The resulting coefficients are in the range of the reported values found in literature.
Large optical annular detectors were realized using polymer optical fibers and a Mach-Zehnder interferometer.
Photoacoustic measurements were performed and compared to numerical simulations. Furthermore, a simple
deconvolution algorithm was developed and applied to reduce artifacts in the images.
For photoacoustic imaging (PAI) so called integrating detectors are used. We developed two types of fiber-based
integrating detectors for photoacoustic tomography (PAT). First images of phantoms with simple structures
reconstructed from data collected with fiber-based detectors are presented.
Photoacoustic Tomography is an emerging imaging technology mainly for medical and biological applications. A
sample is illuminated by a short laser pulse. Depending on the optical properties the electromagnetic radiation
is distributed and absorbed. Thereby local temperature increase generates thermal expansion and broadband
ultrasonic signals, also called photoacoustic signals. Unlike conventional ultrasound in photoacoustic imaging the
contrast depends on the optical properties of the sample which provides not only morphologic information but
also functional information. This way photoacoustic imaging combines the advantages of optical imaging (high
contrast) and ultrasonic imaging (high spatial resolution) and is particularly suited for medical applications like
mammography or skin cancer detection. Our group uses integrating line detectors instead of ultrasonic point
receivers. Line detectors integrate the pressure along one dimension whereby the 3D problem is reduced to a 2D
problem and enables a tomography setup that requires only a single axis of rotation. Implementations of line
detectors use optical interferometers, e.g. a Fabry-Perot interferometer or a Mach-Zehnder interferometer. We
use free-beam interferometers as well as fiber-based interferometers for collecting photoacoustic signals. The latter
are somewhat easier to handle because they require fewer optical components. Finally, the advantages of optical
detection methods over piezoelectric detection methods are the better frequency response and the resistance
against electrical interference from the environment. First measurements on phantoms and image reconstruction
using a time reversal method demonstrated the capability of integrating line detectors for collecting broadband
ultrasonic signals for photoacoustic tomography.
Currently two different types of integrating line sensors are used in photoacoustic tomography (PAT). Thin film
piezoelectric polymer sensors (PVDF) are characterized by compactness, easy handling and the possibility to
manufacture sensing areas with different shape. However, they are vulnerable to electrical disturbance and to scattered
light from the illuminated sample. Also optical sensors are used as integrating line sensors in combination with some
kind of interferometric setup. For example, one arm of a
Mach-Zehnder interferometer or the cavity of a Fabry-Perot
interferometer can be used as line detector. In both cases, the light wave either propagates freely in the liquid or is guided
in an optical fiber. Such sensors are quite immune against noise sources described above and suitable for high bandwidth
detection. One drawback is the limited mobility due to the complex arrangement of the setup.
This study is focused on the comparison of the different implementations of line detectors, mainly on directivity and
sensitivity. Shape and amplitude of signals generated by defined sources are compared among the various sensor types.
While the shape of the signals recorded with the optical free beam detector matches quite well to the simulation the
signals detected with the PVDF detector are affected by directivity effects. This causes a strong distortion of the signal
shape depending on the incident angle of the acoustic wave. How these effects influence the reconstructed projection
image is discussed.
In photoacoustic (also called optoacoustic or thermoacoustic) tomography acoustic pressure waves are generated
by illumination of a semitransparent sample with pulsed electromagnetic radiation. Subsequently the waves
propagate toward the detection surface enclosing the sample. The inverse problem consists of reconstructing the
initial pressure sources from those measurements. By combining the high spatial resolution of ultrasonic imaging
with the high contrast of optical imaging it offers new potentials in medical diagnostics. In certain applications
of photoacoustic imaging one has to deal with media with spatially varying sound velocity, e.g. bones in soft
tissue. These inhomogeneities have a strong influence on the propagation of photoacoustically generated sound
waves. Image reconstruction without any compensation of this effect leads to a poor image quality. It is therefore
essential to develop reconstruction algorithms that take spatially varying sound velocity into account and are
able to reveal small structures in acoustically heterogeneous media. A model-based time reversal reconstruction
method is presented that is capable of reconstructing the initial pressure distribution despite variations of sound
speed. This reconstruction method calculates the time reversed field directly with a second order embedded
boundary method by retransmitting the measured pressure on the detector positions in reversed temporal order.
With numerical simulations the effect of heterogenous media on sound propagation and the consequences for
image reconstruction without compensation are shown. It is demonstrated how time reversal can lead to a
correct reconstruction if the distribution of sound speed is known. Corresponding experiments with phantoms
consisting of areas with spatially varying sound velocity are carried out and the algorithm is applied to the
measured signals.
Photoacoustic imaging is based on the generation of acoustic waves in a semitransparent sample (e.g. soft
tissue) after illumination with short pulses of light or radio waves. The goal is to recover the spatial distribution
of absorbed energy density inside the sample from acoustic pressure signals measured outside the sample
(photoacoustic inverse problem).
If the acoustic pressure outside the illuminated sample is measured with a large-aperture detector, the signal
at a certain time is given by an integral of the generated acoustic pressure distribution over an area that is
determined by the shape of the detector. For example a planar detector measures the projections of the initial
pressure distribution over planes parallel to the detector plane, which is the Radon transform of the initial
pressure distribution. Stable and exact three-dimensional imaging with planar integrating detector requires
measurements in all directions of space and so the receiver plane has to be rotated to cover the entire detection
surface.
We have recently presented a simpler set-up for exact imaging which requires only a single rotation axis and
therefor the fragmentation of the area detector into line detectors perpendicular to the rotation axis. Using a
two-dimensional reconstruction method and applying the inverse two-dimensional Radon transform afterwards
gives an exact reconstruction of the three-dimensional sample with this set-up.
In order to achieve high resolution, a fiber based Fabry-Perot interferometer is used. It is a single mode fiber
with two fiber bragg gratings on both ends of the line detector. Thermal shifts and vibrations are compensated by
frequency locking of the laser. The high resolution and the good performance of this integrating line detector has
been demonstrated by photoacoustic measurements with line grid samples and phantoms using a model-based
time reversal method for image reconstruction. The time reversed pressure field can be calculated directly by
retransmitting the measured pressure on the detector positions in a reversed temporal order.
Photoacoustic imaging is based on the generation of acoustic waves in a semitransparent sample after illumination with
short pulses of light or radio waves. The goal is to recover the spatial distribution of absorbed energy density inside the
sample from acoustic pressure signals measured outside the sample (photoacoustic inverse problem). We have proposed
a numerical method to calculate directly the time reversed field by re-transmitting the measured pressure on the detection
surface in reversed temporal order. This model-based time reversal method can solve the photoacoustic inverse problem
exactly for an arbitrary closed detection surface. Recently we presented a set up which requires a single rotation axis and
line detectors perpendicular to the rotation axis. Using a two-dimensional reconstruction method, such as time reversal in
two dimensions, and applying the inverse two-dimensional radon transform afterwards gives an exact reconstruction of a
three-dimensional sample with this set up. The resolution in photoacoustic imaging is limited by the acoustic bandwidth
and therefore by acoustic attenuation, which can be substantial for high frequencies. This effect is usually ignored in
reconstruction algorithms but has a strong impact on the resolution of small structures. It is demonstrated that the model
based time reversal method allows to partly compensate this effect.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.