Enhanced laser surface ablation with an integrated photoacoustic imaging and high intensity focused ultrasound system

Abstract

Background and objective:

Combined laser and ultrasound treatments have been found to have synergistic effects, which may be of particular note in dermatology. We aim to investigate the potential of this technology for dermatology through in vitro and ex vivo experiments.

Methods:

In vitro tissue phantoms made of agar and tattoo ink and tattooed ex vivo chicken breast tissue were used. An integrated photoacoustic imaging and high intensity focused ultrasound (HIFU) system, using a 5-ns tunable OPO laser system and a 5 MHz HIFU transducer, was used to perform photoacoustic analysis to identify the optical contrast, and perform combined laser and ultrasound ablation. On the tissue phantoms, lines of ablation were created under various operating conditions. The samples were then quantified to determine the level of ablation. Same procedures were performed on the tattooed chicken breast tissue and the tattoo was removed by using combined laser and ultrasound.

Results:

Ablation in the in vitro tissue phantoms was observed with properly synchronized laser and ultrasound while no ablation was found with either laser or ultrasound alone. Increases to the intensity or pulse duration of ultrasound caused an increase in ablation to the samples. The tattoo was removed from the ex vivo chicken breast using combined laser and ultrasound with a radiant exposure of 1.2 J/cm² while laser and ultrasound alone were unable to remove the tattoo.

Conclusions:

We determined that by supplementing nanosecond laser pulses with ultrasound, ablation and tattoo removal can be achieved at laser radiant exposures levels would otherwise be ineffective. The area of ablation can be adjusted through changes in the intensity and duration of the ultrasound burst with a constant laser intensity. Additionally, the system can be used to perform photoacoustic analysis of the tissue to estimate the relative optical absorbance at various available wavelengths, allowing for pretreatment analysis.

Introduction

For the last half of a century lasers have served an important role in medicine. In 1983 Anderson and Parrish introduced selective photothermalysis [1], which was later expanded upon by Altshulur et al. [2]. where short laser pulses are used to destroy optically absorbent targets, without damaging the surrounding tissues. The ability to selectively destroy optically absorbent material has been widely used in dermatology allowing for the removal of skin discolorations [3–5], hair follicles [6, 7], and tattoos [8–12]. Alternatively, by using a laser wavelength that selectively targets the water in the dermis, widespread skin resurfacing can be achieved [13, 14]. More recent developments in laser technology have led to fractional laser resurfacing [15, 16] where narrow channels of deeper ablation are created with spaces of healthy skin between to promote rapid healing.

A 2006 survey found that 24% of men and women between 18 and 50 were found to have tattoos, but 16-21% of those with tattoos had considered getting at least one removed [17]. Generally, laser tattoo removal is expected to require 6-10 treatments, although brighter colors especially yellows and oranges can be especially difficult to treat [10]. Each treatment may cost $200. The most common complication following laser treatments is pigment changes (i.e. hypopigmentation or hyperpigmentation), which are generally more common in people with darker skin [18, 19]. Testing of the treatment on a less visible section of skin is often performed to identify if complications are likely to occur. Lower radiant exposures can be used to prevent pigmentary changes, however the lower exposure level often leads to poorer outcomes and may require more treatments, increasing the cost to the patients. Non-laser techniques have been developed such as high intensity focused ultrasound (HIFU) [20] or radiofrequency (RF) [21]. However, lasers have remained the gold standard. This is in large part because HIFU and RF therapies have been found to be inconsistent and have not demonstrated a clear improvement over laser therapies [22].

In this study we are investigating the potential of using high intensity focused ultrasound to supplement existing laser treatments. Concurrent laser and ultrasound has been shown to cause multiple synergistic effects. Whiteside et al. demonstrated that ultrasound bursts can be used to increase laser penetration, by changing the optical properties of skin [23]. Vangiparum et al. successfully used acoustic shock wave therapy after laser tattoo removal to improve clearance of the tattoo by increasing fluid transfer and lymphatic activity[12]. Jo and Yang found that combined laser and ultrasound treatment causes a dramatic increase in heating [24]. Hu et al. demonstrated that combined laser and ultrasound can be used to selectively target micro-vessels without damaging the surrounding [25]. Similar studies have described how combined laser and ultrasound resulted in increased cavitation [26], which has been used with optically absorbent nanoparticles for targeted cavitation [27, 28]. These previous ablation studies have looked at internal applications, while this study intends to explore dermatological applications.

The ability of combined laser and ultrasound to increase heating and micro-cavitation in tissue could be beneficial when applied to dermatological uses. This would allow for significantly lower radiant exposure levels to be used while still achieving thermolysis or ablation. By using ultrasound intensities that do not directly cause damage, collateral damage outside of the laser illumination can be avoided. Changes to the ultrasound parameters allow for changes in the size or intensity of the ablation while maintaining constant laser settings. Additionally, since ultrasound has excellent penetration through soft tissue and can be focused with HIFU transducers, ultrasound could be focused into the dermal layer to spare the epidermis. This could lead to applications that keep the surface of the skin intact while still targeting the underlying tissue.

The presence of ultrasound transducers could also be used to perform photoacoustic analysis of the targeted tissues to predict the effectiveness and likelihood of adverse reactions. The photoacoustic effect occurs when short laser irradiations cause brief thermal expansions in optically absorbent material. The thermal expansion then mechanically creates an ultrasound wave which can be detected by an ultrasound transducer. The strength of the ultrasound wave is related to the optical absorption of the sample. Detection of these waves is performed through photoacoustic imaging or photoacoustic tomography. The resulting data can be used to create an image based on the optical absorption of the sample, and can be used to identify biological structures with differing optical absorption any available laser wavelength [29, 30]. This technique has been applied in vivo to identify various dermatological pathologies [31], and has been used to identify dyes used in fabrics [32].

When performing selective photothermalysis, optimal treatment is achieved at a wavelength that separates the absorption coefficient of the targeted pigment from that found naturally in the skin [2]. The absorption coefficient naturally found in patient’s skin varies depending on natural skin tone. In the case of tattoo removal, the optical properties of the pigment are likely only generally known [33]. Using ultrasound transducers to perform photoacoustic testing with all available laser wavelengths will reveal which will provide the optimal treatment.

The equation for the generation and propagation of the photoacoustic signal can be seen in (1) as described by Wang and Wu [34]. The generation of the pressure wave can be seen on the left side of the equation, and the propagation of the wave can be seen on the right side.

$$\frac{-\beta }{C_p}\frac{\partial H(r,t)}{\partial t}=\left({\nabla }^2-\frac{1}{v^2}\frac{{\partial }^2}{\partial t^2}\right)p(r,t)$$ (1)

β is thermal coefficient for volume expansion, C_p is the specific heat of the tissue at constant pressure, v is the speed of sound, p(r,t) is the pressure at a particular location and time, and H(r,t) is the heating function at a particular location and time. The heating function is further described in (2)

$$H=\eta {\mu }_a\Phi$$ (2)

Where η is the percentage of absorbed energy that is converted to heat, μ_a is the coefficient of optical absorption, and Φ is the optical fluence rate. When different laser treatments are used on the same sample the only factors expected to change will be in the heating function. The optical fluence rate is primarily determined by the geometry and intensity of the laser irradiation, while η and μ_a are wavelength dependent. By performing preliminary testing at low radiant exposure levels it is possible to compare the changes in η and μ_a without causing any physical effects. The ability to predict the heating function in individual patients may lead to laser wavelength optimizations. The optical properties that determine the strength of photoacoustic signal are the same properties used in selective photothermalysis to preferentially target optically absorbent tissues. This information could be used for both thermal based treatments such as skin resurfacing and in treatments relying on mechanical effects from the thermal expansion such as scar or tattoo removal.

If ultrasound transducers are used to reduce the required laser intensity, a variety of laser system may be available including tunable OPO systems despite their low power outputs. The combination of an increased variety of laser wavelengths and photoacoustic testing could be especially valuable for tattoo removal. An OPO laser system would be able to test the spectra of various pigments in a similar fashion as has been used to identify fabric dyes [32]. The spectra of the tattoo could then be compared to the spectra of the patient to find the optimal treatment wavelength.

In this study we will use a nanosecond laser and ultrasound system to demonstrate combined irradiation to create ablation in agar tissue phantoms. The effects of changing various parameters will be tested to explore the underlying mechanics of the ablation. By performing ablation on the surface of tissue phantom under a variety of treatment parameters, consistent images can be taken and processed for quantitative analysis over a variety of treatment parameters. The techniques explored through the in vitro experiments will be verified using a tattooed ex vivo chicken breast.

Materials and Methods

System

The experimental setup involved the use of a 10-Hz, 532 nm pump laser (Surelight SLI-30, Continuum, Santa Clara, CA) and an OPO system (SLOPO Plus, Continuum, Santa Clara, CA) which was used to provide 5-ns laser pulses with wavelengths between 680-800 nm. The laser system additionally triggered a function generator (33250A, Agilent Technologies, Santa Clara, CA) that was connected to a power amplifier (350L RF Power Amplifier, ENI Technology Inc., Rochester, NY), and then to a 5 MHz spherically focused HIFU transducer (SU-108-013, Sonic Concepts, Bothell, WA). A photoacoustic imaging system, as described in Jo and Yang[35], held the transducer and laser light in alignment. A diagram of the system can be seen in Fig. 1. The coaligned laser and transducer spots were focused on the surface of the sample. Both the laser and ultrasound spots had a gaussian distribution and matching full width half maximum (FWHM) of 0.6 mm at the surface of the sample. A 3D scanning stage (Thorlabs, Newton, NJ) was used to move the system across the sample. The movement of the system was at a speed of 0.5 mm/s. Repeated treatments of any locations on the sample were avoided. A pulse/receiver system (Panametrics NDT, Olympus, Center Valley, PA) was used to ensure correct alignment of the laser and ultrasound.

Figure 1:

A diagram of the photoacoustic imaging/ablation system. The pump laser is used to supply 532 nm light at 10 Hz to the OPO system where it is converted to the selected wavelength. The resulting light is directed into a conical lens and into a condensing lens to bend the light around the transducer to the transducer. The function generator is triggered by the pump laser. The signal is then amplified and delivered to the HIFU transducer.

A delay was added in the function generator to synchronize the arrival of the laser and ultrasound at the surface of the sample. The precise time between laser pulses was 100.1978 ms, and the travel time of the ultrasound wave from the transducer to the sample was 0.0233 ms. A delay of 100.1700 ms was added at the function generator to ensure that the ultrasound burst was in effect at the surface of the sample 4.5 μs before the laser pulse arrived.

During synchronized laser and ultrasound radiation, the trigger from the first laser pulse was used to generate an ultrasound burst which would travel to the sample and synchronize with the second laser pulse. The duration of ultrasound burst was greater than the travel time of the ultrasound wave, meaning the function generator was still generating the signal while the second laser pulse was fired. Our function generator was unable to detect a trigger while generating a signal. As a result, the function generator could only be triggered again by the third laser pulse and the resulted ultrasound burst would synchronized with the fourth laser pulse. Therefore, ultrasound was only applied to every other laser pulse.

The acoustic pressure produced by the HIFU transducer at the focal spot was measured with a standard needle hydrophone (0.5 mm Needle hydrophone SN 1462, Precision Acoustics, Dorchester, UK), and then further simulated through FDTD code developed by Hallaj and Cleveland [36], which was verified by Huang et al. [37]. This calibration allowed for an estimation of the peak pressure from the ultrasound burst at various input voltage levels from the function generator.

Testing was performed in a tank of degassed and deionized water. The position and speed of the stage were controlled by a connected computer. The OPO laser system was primarily used to provide 5 ns pulses of 680 nm light at 10 Hz, which were the optimal operating conditions of the system. The laser and focusing system had a minimum radiant exposure of 1.2 J/cm² during normal operation, which was used as a baseline for most experiments. The baseline ultrasound parameters were 200 μs bursts of 5 MHz ultrasound with a peak negative pressure of 3.6 MPa. The resulting ultrasound duty cycle was 0.01%, preventing an overall temperature increase from ultrasound irradiation. The negative pressure in this study is lower than was used by Zhou and Gau which was up to 15.7 MPa [38]. The radiant exposure levels of the laser are also less than those normally seen in treatments. Izikson et al. compared an alexandrite 755 nm 30-50 ns laser with a 758 nm 500 picosecond laser with minimum radiant exposures of 8 J/cm² and 2.6 J/cm² respectively [9], both of which used greater radiant exposure than 1.2 J/cm² as used in this study. Ross et al. used an Nd:Yag laser with 10 ns pulses that was also operated with a 8 J/cm² radiant exposure [11].

Image Processing

All the experiments involved scanning the system across the tissue phantom in lines using the motorized stage during treatment. The lines were 10 mm long and 2 mm apart from each other. Once the treatment was complete the samples were taken for microscopic imaging. The resulting images were then analyzed and quantified through MatLab (MathWorks, Natick, MA). Quantification was performed by thresholding the images and then calculating the average width of the line of removed material. The threshold was set at the highest level possible that kept the control surfaces between the lines above the threshold. An example of this process is seen in Fig. 2. All trials of any particular test were performed at one time on the same tissue phantom. Multiple images were taken as needed to image all trials. Identical lighting conditions and image processing (including the same threshold values) were used for all images of the same data set.

Figure 2:

An example of the image processing and quantification of results. a) The base image of the sample. b) The image after thresholding. The average with of each line after thresholding was used to quantify the ablation.

Tissue phantoms

The in vitro tissue phantoms were created using a 2% agar (Agar pure powder, ThermoFisher Scientific, Waltham, MA) solution as described by Zell et al. [39]. Tattoo ink (True Black, BZ Ink, Dragon tattoo supply, Dayton, OH ) was added to the solution to provide optical absorption. The solution was poured into a cylindrical mold, which was slightly over filled so that when the sample was slightly domed. The excess material was removed from the top of the sample to create a level surface. The sample was then removed from the mold and flipped over so that the smooth bottom was used as the testing surface.

Experimental Procedure

A series of five experiments were performed in this study. The goal of the experiment 1 was to demonstrate that the photoacoustic system was capable of photoacoustic spectroscopy through the same laser system and ultrasound transducer that would later be used for ablation. Experiments 2-4 examined changes to the level of ablation in tissue phantoms when the synchronization, intensity, and pulse duration were changed independently. The results were imaged and quantified for comparison. In experiment 5 the synergistic effects of combined laser and ultrasound seen in the previous experiments are demonstrated ex vivo by removing a tattoo from chicken breast tissue.

Experiment 1: Photoacoustic spectrometry -

To demonstrate the ability of the photoacoustic system to perform photoacoustic spectroscopy, 5 tissue phantoms were created with 5 different commercially available tattoo inks: true black, cherry bomb (red), teal, dragon green, and lemon yellow (BZ ink, Dragon tattoo supply, Dayton, OH). To create the colored samples 0.06% concentrations of the tattoo dyes were used. The spectra of each of these dyes was determined with a spectrophotometer (Evolution 600, Thermo scientific, Waltham, MA) over the visible and near infrared portions of the spectrum from 400-800 nm. The OPO system was then used to measure the photoacoustic signal generated by each of these sample from a range of 680-800, and the results were compared. Thirty-two photoacoustic measurements were averaged for each wavelength on each sample.

Experiment 2: Synchronization of laser and ultrasound -

The optimal delay time between the start of the ultrasound and laser was examined by varying the delay used in the function generator. For this experiment the laser was operated at 680 nm with a radiant exposure of 1.2 J/cm². The ultrasound burst duration was 200 μs with a peak negative pressure of 3.6 MPa. A variety of synchronizations were tested with the laser pulse arriving before, during, or after the 200 μs ultrasound burst. The goal of this experiment was to examine if direct interaction of the laser pulse and ultrasound burst were required to create a synergistic effect, and to determine what the optimal synchronization between the two should be. For consistency the ultrasound was always triggered by every other laser pulse even when it was possible to trigger on every pulse.

Experiment 3: Intensity testing -

Ablation was also tested using different laser and ultrasound intensities. For this experiment the laser was operated at 680 nm with radiant exposures of 1.2 J/cm² and 1.8 J/cm², which were the lower and upper limits of the our system for radiant exposures. These radiant exposure levels were both well below the 8 J/cm² generally used to represent clinical treatments [9, 11] and were not capable of ablation without supplemental ultrasound. Both radiant exposure levels were tested with a variety of supplemental ultrasound intensities with peak negative pressures ranging up to 4.3 MPa. The goal of this experiment was to identify the threshold of ultrasound intensity to create ablation for each of the laser radiant exposures, and to compare the different levels of ablation at various laser and ultrasound intensities. A constant ultrasound burst duration of 200 μs was used in this experiment.

Experiment 4: Ultrasound burst duration -

Changes in ablation as the ultrasound burst duration was varied were explored. The burst duration was varied from 40-2000 μs by changing the number of cycles in the 5 MHz ultrasound burst. The laser was operated at a 680 nm with a radiant exposure of 1.2 J/cm². The peak negative pressure of the ultrasound burst was 3.6 MPa. Timing was set so that the laser would consistently arrive 4.5 μs into the ultrasound burst at the surface of the sample. This test was used to determine if all portions of the ultrasound burst were responsible for the synergistic surface ablation or if only the portion that overlapped with the laser pulse contributed.

Experiment 5: Ex vivo tattoo removal -

To confirm that tattoo removal could be achieved with the photoacoustic system a black line was tattooed onto chicken breast tissue with a commercial tattoo gun (Dragon tattoo supply, Dayton, OH). A photoacoustic image of the tattoo was taken, and tattoo removal was performed. The laser radiant exposure used for both the imaging and removal was 1.2 J/cm² with 680 nm light. The peak negative pressure of the ultrasound burst was 4.3 MPa and the ultrasound burst length was increased to 4 ms where tattoo removal was observed. The effects of laser alone and ultrasound alone were compared with the combined treatment. The goal of this experiment was to demonstrate that tattoo removal was possible with the lowest radiant exposure of our system when supplemented with ultrasound.

Statistical analysis

The width of each line was calculated by counting pixels in each row of the thresholded images and converting to physical units. The widths of each row in the image were then averaged and presented as mean ± standard deviation. If there was a problem in any region of the image (i.e a crack in the tissue phantom) the same portion was removed from each of the lines for calculations. Differences between two groups was determined using Student’s t-Test. Statistical significance was set at p < 0.05.

Results

Experiment 1:

The spectra from spectrophotometry for the 5 different colored samples (black, teal, green, yellow, and red) are shown in Fig. 3a. The absorbance peaks of the teal, green, yellow, and red samples were 608, 429, 555, and 434 nm respectively. The black dye did not have a peak over the tested wavelengths, but absorbance was highest at 400 nm and gradually dropped off as the wavelength was increased. The wavelengths that were used for the photoacoustic measurements covered the red and near infrared portions of the spectrum from 680 to 800 nm. Fig. 3b–f show the normalized results from photoacoustic spectroscopy and its comparison with spectrophotometry over the range used for photoacoustic spectroscopy. As expected the teal and green samples had a stronger photoacoustic signal near the peaks identified by spectrophotometry. The red and yellow samples had very little change in the 680-800 nm range due to the low absorption predicted by spectrophotometry. These results demonstrate that photoacoustic spectroscopy can be performed using the same laser and ultrasound transducer that will be used for ablaiton. There are some differences in the comparisons between the photoacoustic measurements and the spectrophotometry results. This is likely because spectrophotometry is effected by both light scattering and absorptions in the medium, while the photoacoustic signal is largly determined by absorptions as well as the Grüneisen constant. Photoacoustic testing may be used when the absorption peaks of the dyes are unknown, or it may be used to compare the relative absorption bewteen the targeted tissue and the surrounding tissue at various wavelengths. During the course of photoacoustic measurements no ablaiton was seen in the samples, this suggests that every other laser pulse that is out of sync with the ultrasound burst would be ineffective.

Figure 3:

Experiment 1 - a) The absorbance spectra, using spectrophotometry from 400-800 nm, of five tissue phantoms colored with different dyes black, teal, green, red, and yellow. The peak wavelength for each of the samples: black –400 nm, teal – 608 nm, green – 429, red – 555, yellow – 434. b-f) The normalized results of the photoacoustic spectra are compared to the spectrophotometry results in the 680-800 nm range for each of the dyes. b) black, c) teal, d) green, e) red, and f) yellow. Error bars represent standard deviations of 32 measurements.

Experiment 2:

Synchronization testing was performed to determine the optimal arrival time of the laser pulse relative to the ultrasound burst. The results can be seen in Fig. 4. All times in the plot are based on the surface of the sample accounting for travel time of the ultrasound burst, and were verified with an oscilloscope. When the laser arrived at the surface of the sample 34.5 μs or more before the arrival of the ultrasound burst, only a small amount of ablation was seen. When the laser arrived within 24.5 μs of the ultrasound burst the width of ablation was similar to when the laser arrived during the beginning of the ultrasound. The width of ablation dropped as the laser arrived further into the ultrasound burst until no ablation was seen. When the laser arrived after the ultrasound no ablation was seen.

Figure 4:

Experiment 2 - The results of changing the synchronization between the arrival of the laser pulse and ultrasound burst at the surface of the tissue phantom. Negative times represent the laser pulse arriving before the ultrasound burst. The shaded region indicates the time where the 200 μs ultrasound burst was in effect. Error bars represent standard deviations of 150 width measurements of each line of ablation.

The similarities in ablation between trials where the laser arrived during the beginning of the pulse and just before the pulse suggests that the laser pulse is causing an effect on the sample that lasts longer than would be expected for the pressure wave in a photoacoustic signal. The creation of nano- or micro-bubbles through cavitation from the laser pulse is one possible reason that the laser arriving before the ultrasound burst can still enhance ablation, because the residual cavitation may be caught up by the subsequent ultrasound burst. The lifetime of the laser induced bubbles would then determine the length of time that the laser could arrive before the ultrasound burst to enhance ablation, which may explain the reduction in ablation as the laser pulse arrives further ahead of the ultrasound burst, as fewer bubbles survived until the arrival of the ultrasound burst. These results provide further evidence that every other laser pulse that was not supplemented with ultrasound should provide negligible effect compared to the properly synchronized pulses.

Experiment 3:

Fig. 5 demonstrates the changes in ablation for two different radiant exposure levels (1.2 J/cm² and 1.8 J/cm²) as the intensity of the ultrasound was increased. A minimum ultrasound threshold to cause ablation was noted in each case. Any trials with an ultrasound pressure below that threshold did not result in ablation. When the laser radiant exposure was 1.2 J/cm², ablation was seen when the ultrasound had a peak negative pressure equal or greater than 2.8 MPa. When the radiant exposure was raised to 1.8 J/cm², ablation was seen starting at 1.9 MPa. Overall the width of ablation was greater when either the radiant exposure of the laser or the magnitude of the ultrasound pressure was increased. Greater ablation was seen in all trials with 1.8 J/cm² radiant exposure compared to trials with 1.2 J/cm² radiant exposure with the same ultrasound intensity.

Figure 5:

Experiment 3 - The results of various ultrasound intensities with two different radiant exposures. Both radiant exposures displayed a threshold pressure that was required to cause ablation. The higher laser radiant exposure decreased the required pressure threshold. It is additionally seen that be increasing the intensity of the ultrasound pressure, the level of ablation is increased. Error bars represent standard deviations of 400 width measurements of each line of ablation.

Experiment 4:

The effect of the ultrasound burst duration is presented in Fig. 6. It is clearly seen that the ultrasound burst duration had a direct impact on the area of ablation. The increase in ablation was much more noticeable at the shorter durations than the longer durations, indicating that there may be an upper limit to the increase in ablation that can be caused by increasing the ultrasound burst duration. These results support the findings in the synchronization testing, where the width of ablation gradually decreased as the laser pulse arrived later in the ultrasound burst.

Figure 6:

Experiment 4 - The results of various ultrasound burst lengths. As the pulse lengthened the width of ablation increased, indicating that ablation occurred throughout the whole ultrasound burst not just at the portion that overlapped with the laser. Error bars represent standard deviations of 400 width measurements of each line of ablation.

Experiment 5:

The synergistic effects from combined laser and ultrasound seen in the in vitro results were used to remove tattoo ink from an ex vivo chicken breast. Photoacoustic imaging was performed on the tattoo and is presented in Fig. 7a. The average photoacoustic signal of each pass across the tattoo is presented in Fig. 7b where a significant increase in photoacoustic signal is seen in the tattoo compared to the surrounding tissue. The tattoo was treated with ultrasound alone, laser alone and a combined treatment in Fig. 7c. The ultrasound alone did clear the tattoo ink that had stained the surface but was ineffective at removing the deeper ink. The laser treatment alone had no effect on the tattoo. The combined treatment was able to remove the tattoo completely.

Figure 7:

Experiment 5 - a) Maximum amplitude projection (MAP) of a photoacoustic image taken of a tattooed ex vivo chicken breast. B) Average photoacoustic signal from each of the passes across a sample in the photoacoustic image. c) An image of the sample after applied treatments from the ultrasound alone (US), the laser alone(L), and the combined laser ultrasound treatment (C).

Discussion

The results of this study show that our integrated photoacoustic imaging and HIFU system can be used to supplement laser surface ablation and allows for photoacoustic analysis of the targeted tissues. It is possible to use laser and ultrasound intensity that independently are unable to cause damage, but in conjunction the surface of the tissue phantom can be removed. The radiant exposure of 1.2-1.8 J/cm² used in this study with supplemental ultrasound was able to cause ablation and perform tattoo removal on an ex vivo sample, compared to other studies where 8 J/cm² was used to represent clinical treatments [9, 11]. It is expected that as the exposure level of the laser is reduced, higher ultrasound intensities and pulse durations could be used to compensate. However, the selectivity of the laser treatment would be expected to be reduced.

Our study also shows that combined ultrasound and laser ablation is not simply the result of the combined pressure waves. If the combined pressure wave from the laser and ultrasound were solely responsible for the ablation, then changes to the length of the ultrasound burst would not be expected to enhance the level of ablation. Cavitation as described by Cui et al. [26], is most likely the cause of the tissue phantom surface removal that was observed in this study, although this study demonstrated that the a synergistic effect is seen even when the laser and ultrasound are slightly out of synchronization. If cavitation was initiated by the laser, then the ultrasound would be able to drive the bubbles to cause further damage to the sample. Using ultrasound driven bubbles to cause ablation has been used in treatments where micro bubbles are injected into the blood stream and driven externally by ultrasound [40]. The demonstration that changes in ultrasound pulse duration and intensity can dramatically impact the level of ablation, may have impact outside of tattoo removal and dermatology. The phenomenon seen in this study may apply to the other treatments using combined laser and ultrasound.

It should be noted that only one laser configuration was used in this study which uses and tunable OPO system which is not representative of clinically available systems. However, it is expected that ultrasound will provide similar synergistic benefits to most other laser sources, including continuous lasers, picosecond lasers, fractional lasers or future laser developments. Future work will need to be performed to replicate these results in vivo, and to examine the biological response to combined laser and ultrasound ablation. Additional studies are needed to directly compare the effectiveness of combined laser and ultrasound treatments with more traditional laser therapies.

Conclusion

This study demonstrates that when both laser and ultrasound are used simultaneously, ablation can be enhanced when compared to laser-alone or ultrasound-alone. Changes to the parameters of the ultrasound burst can be used to easily adjust the level of ablation while maintaining constant laser parameters. Additionally, the introduction of ultrasound transducers allows for photoacoustic analysis which can be used for pretreatment estimates of the response in the targeted tissue from laser treatments. The benefits from supplemental ultrasound to existing laser-based therapies may be beneficial in dermatology and other medical fields.