The detection of ultrasound in photoacoustic tomography (PAT) usually relies on ultrasonic transducers in contact with the biological tissue through a coupling medium. This is a major drawback for important potential applications such as surgery. Here we report the use of a remote optical method, derived from industrial laser-ultrasonics, to detect ultrasound in tissues. This approach enables non-contact PAT (NCPAT) without exceeding laser exposure safety limits. The sensitivity of the method is based on the use of suitably shaped detection laser pulses and a confocal Fabry-Perot interferometer in differential configuration. Reliable image reconstruction is obtained by measuring remotely the surface profile of the tissue with an optical coherence tomography system. The proposed method also allows non-contact ultrasound imaging (US) by applying a second reconstruction algorithm to the data acquired for NCPAT. Endogenous and exogenous inclusions exhibiting optical and acoustic contrasts were detected ex vivo in chicken breast and calf brain specimens. Inclusions down to 0.3 mm in size were detected at depths exceeding 1 cm. The method could expand the scope of photoacoustic and US to in-vivo biomedical applications where contact is impractical.
Photoacoustic tomography (PAT) and ultrasonography (US) of biological tissues usually rely on ultrasonic transducers
for the detection of ultrasound. For an optimum sensitivity, transducers require a physical contact with the tissue using a
coupling fluid (water or gel). Such a contact is a major drawback in important potential applications such as surgical
procedures on human beings and small animal imaging in research laboratories. On the other hand, laser ultrasonics (LU)
is a well established optical technique for the non-contact generation and detection of ultrasound in industrial materials.
In this paper, the remote optical detection scheme used in industrial LU is adapted to allow the detection of ultrasound in
biological tissues while remaining below laser exposure safety limits. Both non-contact PAT (NCPAT) and non-contact
US (NCUS) are considered experimentally using a high-power single-frequency detection laser emitting suitably shaped
pulses and a confocal Fabry-Perot interferometer in differential configuration. It is shown that an acceptable sensitivity is
obtained while remaining below the maximum permissible exposure (MPE) of biological tissues. Results were obtained
ex vivo on chicken breast specimens with embedded inclusions simulating blood vessels optical properties. Sub-mm
inclusions are readily detected at depths approaching 1 cm. The method is expected to be applicable to living tissues.
Photoacoustic tomography (PAT) and ultrasonography (US) of biological tissues usually rely on transducer arrays for the
detection of ultrasound. Obtaining the best sensitivity requires a physical contact with the tissue using an intermediate
coupling fluid (water or gel). This type of contact is a major drawback for several applications such as neurosurgery.
Laser-ultrasonics is an established optical technique for the non-contact generation and detection of ultrasound in
industrial materials. In this paper, the non-contact detection scheme used in laser-ultrasonics is adapted to allow probing
of ultrasound in biological tissues while remaining below laser exposure safety limits. Both non-contact PAT (NCPAT)
and non-contact US (NCUS) are demonstrated experimentally using a single-frequency detection laser emitting suitably
shaped pulses and a confocal Fabry-Perot interferometer. It is shown that an acceptable sensitivity is obtained while
remaining below the maximum permissible exposure (MPE) of biological tissues. Results obtained ex vivo with a calf
brain specimen show that sub-mm endogenous and exogenous inclusions can be detected at depths exceeding 1 cm.
When fully developed, the technique could be a unique diagnostic tool in neurosurgery providing deep imaging of blood
vessels, blood clots and blood oxygenation.
The advantages and limitations of using a beating heart model in the development of intravascular optical coherence tomography are discussed. The model fills the gap between bench experiments, performed on phantoms and excised arteries, and whole animal in-vivo preparations. The beating heart model is stable for many hours, allowing for extended measurement times and multiple imaging sessions under in-vivo conditions without the complications of maintaining whole-animal preparation. The perfusate supplying the heart with nutrients can be switched between light scattering blood to a nonscattering perfusate to allow the optical system to be optimized without the need of an efficient blood displacement strategy. Direct access to the coronary vessels means that there is no need for x-ray fluoroscopic guidance of the catheter to the heart, as is the case in whole animal preparation. The model proves to be a valuable asset in the development of our intravascular optical coherence tomography technology.
Ultrasound-modulated optical tomography (UOT) combines the spatial resolution of ultrasonic waves and the
spectroscopic properties of light to detect optically absorbing and/or scattering objects in highly scattering media. In this
work, a double-pass confocal Fabry-Perot interferometer is used as a bandpass filter to selectively detect the ultrasoundtagged
photons. The limited etendue of the confocal Fabry-Perot interferometer is compensated by using a singlefrequency
laser emitting high-peak-power optical pulses. Compared to photoacoustic tomography, UOT is not only
sensitive to optical absorption but also to scattering properties. In this paper, we consider the detection of absorbing and
scattering objects embedded in thick (30 to 60 mm) tissue-mimicking phantoms and biological tissues. The experimental
evaluation of the spatial resolution of the technique is compared to that expected from the ultrasonic beam intensity
profile. Preliminary results indicate that the edge spread function is influenced by the level of absorption of the
embedded object.
Ultrasound-modulated optical imaging is an emerging biodiagnostic technique which provides the optical spectroscopic
signature and the spatial localization of an optically absorbing object embedded in a strongly scattering medium. The
transverse resolution of the technique is determined by the lateral extent of ultrasound beam focal zone while the axial
resolution is obtained by using short ultrasound pulses. The practical application of this technique is presently limited by
its poor sensitivity. Moreover, any method to enhance the
signal-to-noise ratio must satisfy the biomedical safety limits.
In this paper, we propose to use a pulsed single-frequency laser source to raise the optical peak power applied to the
scattering medium and to collect more ultrasonically tagged photons. Such a laser source allows illuminating the tissues
mainly during the transit time of the ultrasonic wave. A
single-frequency Nd:YAG laser emitting 500-μs pulses with a
peak power superior to 100 W was used. Tagged photons were detected with a GaAs photorefractive interferometer
characterized by a large optical etendue. When pumped by high intensity laser pulses, such an interferometer provides
the fast response time essential to obtain an apparatus insensitive to the speckle decorrelation encountered in biomedical
applications. Consequently, the combination of a large-etendue photorefractive interferometer with a high-power pulsed
laser could allow obtaining both the sensitivity and the fast response time necessary for biomedical applications.
Measurements performed in 30- and 60-mm thick optical phantoms made of titanium dioxide particles dispersed in
sunflower oil are presented. Results obtained in 30- and 60-mm thick chicken breast samples are also reported.
Ultrasound-modulated optical imaging combines the good spatial resolution of ultrasonic waves (mm scale) and the
spectroscopic properties of light to detect optically absorbing objects inside thick (cm scale) highly scattering media.
Light propagating in a scattering medium can interact with an ultrasonic wave thereby being tagged by a frequency shift
equal to the ultrasound frequency or its harmonics. In this paper, a confocal Fabry-Perot interferometer (CFPI) is used as
a tunable spectral filter to detect selectively the ultrasound-tagged photons. The CFPI allows obtaining high spectral
resolution (MHz scale) while maintaining a high light gathering power when compared to other spectroscopic devices of
comparable resolution. The contrast between the tagged photons and the untagged photons can be further enhanced by
cascading CFPI. Moreover, the fast response of the CFPI allows performing measurements within the speckle
decorrelation time typically encountered in biomedical applications. In this paper, the use of a single-frequency laser
emitting powerful optical pulses allows illuminating the scattering medium only during the transit time of the probing
ultrasonic pulses. Consequently, the acoustic and the optical power are both concentrated in time to enhance the signal-to-
noise ratio of the technique while remaining below the biomedical safety limits. The detection of optically absorbing
objects (mm size) inside 30- and 60-mm thick scattering media is presented.
Acousto-optical imaging is an emerging biodiagnostic technique which provides an optical spectroscopic signature and a
spatial localization of an optically absorbing target embedded in a strongly scattering medium. The transverse resolution
of the technique is determined by the lateral extent of ultrasound beam focal zone while the axial resolution is obtained
by using short ultrasound pulses. Although very promising for medical diagnostic, the practical application of this
technique is presently limited by its poor sensitivity. Moreover, any method to enhance the signal-to-noise ratio must
obviously satisfy the in vivo safety limits regarding the acceptable power level of both the ultrasonic pressure wave and
the laser beam. In this paper, we propose to improve the sensitivity by using a pulsed single-frequency laser source to
raise the optical peak power applied to the scattering medium and to collect more ultrasonically tagged photons. Such a
laser source also allows illuminating the tissues mainly during the transit time of the ultrasonic wave to maintain the
average optical power below the maximum permissible exposure. In our experiment, a single-frequency Nd:YAG laser
emitting 500-μs pulses with a peak power superior to 100 W was used. Photons were tagged in few-cm thick optical
phantoms with tone bursts generated by an ultrasonic transducer. Tagged photons were detected with a GaAs
photorefractive interferometer characterized by a large optical etendue to process simultaneously a large number of
speckle grains. When pumped by high intensity laser pulses, such an interferometer also provides the fast response time
essential to obtain an apparatus insensitive to the speckle decorrelation due to mechanical vibrations or tissues
movements. The use of a powerful long pulse laser appears promising to enhance the signal level in ultrasound
modulated optical imaging. When combined with a photorefractive interferometer of large optical etendue, such a source
could allow obtaining both the sensitivity and the fast response time necessary for biodiagnostic applications.
Acousto-optical imaging is based on the detection of strongly scattered light which is in part modulated by its interaction
with an ultrasonic wave. This method benefits from the acoustic uniformity (low acoustic scattering and absorption) of
an optically diffusive medium and the spectrally selective absorption of photons. In this work, we consider the use a
pulsed single-frequency laser to increase the instantaneous optical power applied to the diffusive medium while
maintaining the average power below the maximum permissible exposure. Such a laser source concentrates the
illumination of the diffusive medium during the transit time of the ultrasonic toneburst. This allows collecting more
ultrasound-modulated photons for a given ultrasonic wave amplitude. We found, however, that a pulsed laser of this kind
generates additional noise which limits the sensitivity gain expected from its high peak power. Progress toward sensitive
imaging was achieved by developing methods to reduce the impact of this additional noise. Results obtained with
differential detection, laser beam spatio-temporal homogenization and variable delay synchronization are presented.
With such measures, the use of a pulsed laser appears a promising solution for enhancing the sensitivity in acousto-optical
imaging.
We developed optical tissue phantoms with a novel combination of matrix and scatterers. These phantoms have a well
known scattering microstructure of monodisperse silica microspheres, embedded in elastic silicone. We characterize
their mechanical properties and, some of their optical properties. We also validate the control over the density of
scatterers achieved with our proposed fabrication technique. The properties obtained are a practical combination of
deformability, durability and simplicity of the microstructure. These are illustrated by results on speckle statistics in
optical coherence tomography.
Speckle is always present in Optical Coherence Tomography (OCT) measurements. To a first approximation, the speckle size is determined by the OCT resolution length and the point spread function of the focusing optics in the sample arm. But the speckle size is also affected by the tissue microstructure. We demonstrate this phenomena by performing measurements on optical phantoms with a controlled density of scatterers using time-domain OCT. In the very low density limit, the scatterers are easily identified on the OCT cross-section and, in fact, one can hardly speak of a speckle pattern. The corresponding speckle size is the resolution length axially and the point spread function of the focusing optics transversally. As the number of scatterers increases, a true speckle field appears and the measured speckle size decreases. In the high density limit, the speckle size reaches an asymptotic value that is about 70% of its low-density regime values. In addition to experimental results, theoretical estimates of the limiting speckle size values are presented. Our work contributes to a better understanding of speckle in optical coherence tomography.
As in conventional time-domain optical coherence tomography (OCT), speckle is inherent to any Optical Fourier
Domain Imaging (OFDI) of biological tissue. OFDI is also known as swept-source OCT (SS-OCT). The axial
speckle size is mainly determined by the OCT resolution length and the transverse speckle size by the focusing
optics illuminating the sample. There is also a contribution from the sample related to the number of scatterers
contained within the probed volume. In the OFDI data processing, there is some liberty in selecting the range of
wavelengths used and this allows variation in the OCT resolution length. Consequently the probed volume can
be varied. By performing measurements on an optical phantom with a controlled density of discrete scatterers
and by changing the probed volume with different range of wavelengths in the OFDI data processing, there is an
obvious change in the axial speckle size, but we show that there is also a less obvious variation in the transverse
speckle size. This work contributes to a better understanding of speckle in OCT.
Speckle is inherent to any Optical Coherence Tomography (OCT) imaging of biological tissue. It is often seen as
degrading the signal, but it also carries information about the tissue microstructure. One parameter of interest is
the speckle size. We study the variations in speckle size on optical phantoms with different density of scatterers.
Phantoms are fabricated with a new approach by introducing silica microspheres in a curing silicon matrix,
providing phantoms with a controlled density of scatterers. These phantoms are also solid, deformable, and
conservable. Experimental results are obtained with Time-Domain OCT (TD-OCT). Modeling is performed
by simulating a phantom as a random distribution with of discrete scatterers. Both experimental results and
modeling show that the speckle size varies when there are few scatterers contained within the probed volume,
the latter being defined by the coherence length and the spot size of the focusing optics. As a criterion to
differentiate tissues, the speckle size has the same sensitivity as the contrast parameter that is studied in Ref. 1.
This work also contributes to a better understanding of speckle in OCT.
We propose a robust and efficient delay line using an ensemble of rotating rhombic prisms. Delay lines relying on
rotating elements provide fast and stable operation. Optical systems using rhombic prisms are quite easy to align
since these prisms are efficient even when slightly misaligned. Optical delay lines with a single rotating element
usually have a poor duty cycle and show large nonlinearity in the variation of the optical path lengh with the
angular position. Our delay line improves over existing technology by using off-centroid rotation and reinjection.
Off-centroid rotation allows the use of multiple prisms and, by optimizing the conditions of operation, the duty
cycle is increased and the nonlinearity is decreased. The duty cycle and repetition rate are further increased by
reinjecting the incoming ray towards the delay line when it is not first intercepted by the prism ensemble. We
have designed and built such a delay line using five prisms. The experimental device was tested at 2000 delay
scans per second and provided a duty cycle larger than 80% with about 5% nonlinearity. Higher delay scan rates
are easily achievable with this technology. The delay line was introduced in a time-domain optical coherence
tomography system and example of imaging of biological tissue is provided.
An overview of the activities in low-coherence interferometry (LCI) and optical coherence tomography (OCT) at the Industrial Materials Institute are presented. An innovative optical delay line using rhombic prisms is described. A few industrial applications are described: volume loss in a wear test, combination of LCI with laser-induced breakdown spectroscopy, and modification of an existing rheometer to increase its precision. Preliminary results related to the use of speckle to differentiate tissues and materials in OCT are presented. The speckle dimension is shown to be sensitive to low density of scatterers. An additional parameter extracted from the autocorrelation of speckle is also presented.
A laser-ultrasonic technique is described to non-destructively determine residual stresses in metals such as those produced by shot peening. The method is based on monitoring the small ultrasonic velocity change of the laser-generated surface skimming longitudinal wave (LSSLW) propagating just below the surface. The main advantage of using LSSLW is that the effect of surface roughness induced by shot peening is greatly reduced compared to using surface acoustic waves (SAW). To improve resolution in the measurement of small velocity changes, a cross-correlation technique is used with a reference signal taken on the same but unstressed material in similar conditions. Also, the low-frequency SAW can be used to correct the LSSLW results when affected by minute changes in the path length during the measurements. The validity of the approach is demonstrated by measuring quantitatively the near surface stress in a four-point bending experiment with different levels of surface roughness. Then, scanning results on properly and improperly laser shock peened samples are reported. In particular, the LSSLW velocity variations for the properly peened samples clearly show an increase in the laser-peened area well indicative of a compressive stress.
Daniel Levesque, M. Massabki, Marc Choquet, Christian Neron, N. Bellinger, David Forsyth, C. Chapman, Ronald Gould, Jerzy Komorowski, Jean-Pierre Monchalin
Corrosion has been recognized as a serious problem in the maintenance of aging aircraft. The Industrial Materials Institute (IMI) has explored the use of laser-ultrasonics for the detection of hidden corrosion in metallic lap joint structures. For inspection with painted surfaces, IMI has shown that a resonance spectroscopy approach using a simple two-layer model can be used to determine the thickness of the paint layer and of the top metal skin. Validation of the model has been made using a test sample with a broad range of paint thickness. Once combined with a numerical inversion method, the model is used to produce a thickness map of the top metal skin from measured resonance frequencies. Results from standard samples with flat-bottom holes showed that the laser-ultrasonic technique could detect metal loss below 1%. The reliability of the method was also demonstrated on accelerated corrosion samples. Comparison to X-ray images showed that the laser-ultrasonic method presented a thickness map that had the same accuracy as the X-ray system without the need for dismantling the sample. These results indicated that laser-ultrasonics could be a useful tool not only to inspect aircraft during routine maintenance but also to provide valuable data in the study of corrosion inception and growth in lap joint structures.
In this paper we explore laser induced breakdown spectroscopy (LIBS) at relatively low energies in the range 10 -
350 tJ. We present measurements ofthe threshold laser energy needed for LIBS and the scaling of plasma size and crater
size with energy. The effects of the laser pulse length and gating of the detector on the LIB spectra are studied and we also
assess the use ofmicrojoule LIBS for the identification ofAl alloys.
An improved phase demodulator for the detection of ultrasound based on two wave mixing (TWM) in semi-insulating
photorefractive GaAs crystal is presented. A new optical layout is proposed in which the total laser power is injected into the
crystal to reduce the grating buildup time. The device is then less sensitive to ambient vibrations or motion of the inspected
part. To reduce the sensitivity of the device to amplitude fluctuations, a balanced receiver with a large etendue was
developed. The measured rejection ratio of this balanced receiver is 50 dB. However, in this new optical configuration the
signal and pump beam paths cannot be made equal which results in some sensitivity to high frequency laser phase noise. In
this paper the results and performances of this novel configuration operating with CW and pulsed lasers are discussed. A
comparison with the conventional confocal Fabry-Perot demodulator is also presented.
We present and describe different techniques based on the photorefractive effect that are used for the optical detection of ultrasonic signals. These techniques use the wavefront adaptation properties of the photorefractive effect. They are: the photorefractive beam combiner, the double phase conjugate heterodyne detection and the adaptive photodetector based on non steady state photoelectromotive force. Their respective advantages and drawbacks, are overviewed. We insist on the latest development and performances obtained with the photorefractive beam combiner that seems to use the most promising technique for the detection of ultrasonic signals on rough surfaces. We show that a sensor with near optimum sensitivity can be developed with the same photorefractive crystal at different wavelength in the range of 1 micrometer to 1.55 micrometer.
We present and describe two techniques used for optical detection of ultrasonic signals based on the photorefractive effect. These techniques used the wavefront adaptation properties of the photorefractive effect. In the photorefractive beam combiner, a local oscillator matched to the signal wavefront is created, leading to an homodyne detection system having a large etendue. In the double phase conjugate heterodyne detection system the signal beam wavefront is cleaned by a double phase conjugate mirror and transformed in a plane wave that is sent on a classical heterodyne detection system. Both systems are characterized and used to detect ultrasound.
Optical absorption is obviously of prime interest in the efficiency of laser generation of ultrasound in graphite-epoxy laminates. However, no quantitative spectrum of optical absorption in this composite material has yet been published in the literature. Transmission techniques are inefficient, and other techniques, like attenuated total reflectance or diffusive reflectance, do not give absolute values. The Fourier transform photoacoustic spectroscopy technique seems to be a good alternative that can analyze adequately and quantitatively a graphite-epoxy laminate. We used three different methods to compute the absolute optical absorption from the photoacoustic signal. The three methods are: the saturation of the real part of the photoacoustic spectrum, the comparison of the spectra obtained with two different mirror velocities, and the calibration of the photoacoustic cell with a transmission measurement. The spectra obtained in the IR band of 2.5 to 25 μm are presented, and the problems and limitations of each method are discussed. The results permit a better understanding of the absorption process in the composite laminate, and in this way, will help us enhance the efficiency of laser generation of ultrasound in graphite epoxy.
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.