Gas-filled nanobubbles (NBs) have been used for contrast enhanced ultrasound (US) molecular imaging similar to their microbubble (MB) counterparts. In this work we explore the use of NBs as cancer therapeutic vehicles to enhance radiation therapy and examine the role of photoacoustic (PA) imaging for quantifying the tissue response. Experiments were conducted in 58 mice bearing prostate cancer tumors. The treatment groups consisted of untreated controls (n=14), 8Gy radiation (n=14), US+MB (n=8), US+NB (n=5), US+MB+8Gy (n=12) and US+NB+8Gy (n=5). PA imaging was conducted using the VevoLAZR 21 MHz probe at pre-treatment, 2h and 24h post-treatment. To examine the changes in the tumor microenvironment post-treatment, we measured the tumor oxygen saturation (sO2), total hemoglobin (HbT) and the spectral slope (SS) as a surrogate measurement of vessel size. Histology measures of tumor vascularity/vessel size (CD31) and cellular death (TUNEL) allowed us to interpret the PA biomarker changes. The PA biomarkers revealed the following changes for the US+NB+8Gy group: (a) a 20% decrease in tumor sO2, (b) a 30% increase in tHb and (c) 20% decrease SS. These results suggest that NBs induce vessel damage from US-induced cavitation. Histology elucidates these findings: CD31 staining measuring vascularity decreases by 80% at 24h–this causes the drop in sO2 and increase in HbT; the size of vessels decreases by 10%–this is captured by the SS change. This work shows that PA imaging biomarkers can quantify these vascular changes which in turn enhance the radiation-induced cell death (up to 40%, as measured by TUNEL staining).
In this work, we present the first time use of photoacoustic imaging for assessing the quality of donor kidneys pre-transplantation. There is a pressing clinical need to quantify the fibrotic (scarring) burden in a non-invasive manner to give clinicians crucial information before they decide whether to accept a donor kidney. Our results in human kidneys show that photoacoustic imaging can be a robust tool for assessing the degree of scarring, an important predictor of post-transplantation clinical outcome.
In this work, we examine the potential of photoacoustic imaging for understanding the biophysical mechanism of nanobubble and microbubble-based vascular disrupting therapies. We present for the first time, a direct in-vivo comparison between sub-micron nanobubbles and the commercially available microbubbles. Our results show that PA imaging of tumor oxygenation is capable of measuring the nanobubble-induced almost 40% cell death as a result of vascular disruption.
Solid tumors are typically supplied nutrients by a network of irregular blood vessels. By targeting these vascular networks, it might be possible to hinder cancer growth and metastasis. Vascular disrupting agents induce intertumoral hemorrhaging, making photoacoustic (PA) imaging well positioned to detect bleeding due to its sensitivity to hemoglobin and its various states. We introduce a fractal-based numerical model of intertumoral hemorrhaging to simulate the PA signals from disrupted tumor blood vessels. The fractal model uses bifurcated cylinders to represent vascular trees. To mimic bleeding from blood vessels, hemoglobin diffusion from microvessels was simulated. In the simulations, the PA signals were detected by a linear array transducer (30 MHz center frequency) of four different vascular trees. The power spectrum of each beamformed PA signal was computed and fitted to a straight line within the −6-dB bandwidth of the receiving transducer. The spectral slope and midband fit (MBF) based on the fit decreased by 0.11 dB / MHz and 2.12 dB, respectively, 1 h post bleeding, while the y-intercept increased by 1.21 dB. The results suggest that spectral PA analysis can be used to measure changes in the concentration and spatial distribution of hemoglobin in tissue without the need to resolve individual vessels. The simulations support the feasibility of using PA imaging and spectral analysis in cancer treatment monitoring by detecting microvessel disruption.
In this work, speckle in acoustic-resolution photoacoustic (PA) imaging systems is discussed. Simulations and experiments were used to demonstrate that PA speckle carries structural information related to sub-resolution absorbers.
Numerical simulations of phantoms containing spherical absorbers were performed using Green’s function solutions to the PA wave equation. A 21 MHz linear array was simulated (256 elements, 75×165 µm resolution, bandwidth 9-33 MHz) and used to record, bandlimit and beamform the generated PA signals. The effects of absorber size (10-270 µm) and concentration (10-1000/mm3) on PA speckle were examined using envelope statistics and radiofrequency spectroscopy techniques. To examine PA speckle experimentally, a VevoLAZR system was used to image gelatin phantoms containing 3 and 15µm polystyrene beads, a tissue mimicking radial artery phantom, and murine tumour vasculature in vivo.
Fully developed speckle, as assessed by Rayleigh distribution fits to PA signal envelopes, was present in all images (simulated and experimental) containing at least 10 absorbers per resolution volume, irrespective of absorber size. Changes in absorber size could be detected using the spectral slope of the normalized power spectrum (4.5x decrease for an 80 µm increase in size). PA images of flowing blood in the radial artery phantom also revealed the presence of speckle with intensity that fluctuated periodically with beat rate (4 dB per cycle). Speckle was ubiquitous to all murine tumor vasculature images. During treatment-induced vascular hemorrhaging, the spectral slope decreases by 80% compared to untreated mice. These results demonstrate that photoacoustic speckle encodes information about the underlying absorber distribution.
Photoacoustic (PA) signals carry information of the absorbing chromophores and the light distribution in imaged samples. The dependence of light distribution with optical wavelength affects the accuracy in PA chromophore quantification. Oxygen saturation (sO2) estimations maybe inaccurate in-depth due to the lack of proper fluence compensation. We propose the use of the PA radiofrequency spectral slope (SS) to generate a frequency filter to match the fluence across optical wavelengths.
The SS is calculated from the ratio of the radiofrequency power spectra at the selected optical wavelengths. The SS relays information about the absorbers’ size and the light distribution. At the imaged optical wavelengths of the same sample, the SS-estimated size should in principle remain unchanged. This suggests that any changes in the measured SS as a function of optical wavelength can be attributed to the light distribution. A frequency filter can be designed from the computed SS and applied to compensate the PA images.
A 5mm phantom consisting of fresh blood, intralipid and gelatin was imaged using the VevoLAZR system at 750 and 850nm. A square sliding window sized 1.6mm with 80% overlap is applied to segment the generated radiofrequency signals. The designed ultrasound filter was applied to each segmented signal.
As a result, the fluence-induced depth fluctuations in the sO2 estimations dropped from 9.49%/mm to 1.83%/mm. This will allow for more accurate sO2 estimates that are less depth dependent. The approach provides a new perspective for fluence compensation which can aid in improving chromophore quantification using PA imaging.
Spectral analysis of photoacoustic (PA) signals in the ultrasound frequency domain is a method that analyzes the power spectrum of PA signals to quantify tissue microstructures. PA spectral analysis has been correlated to changes in the size, morphology and concentration of absorbers that are smaller than the system spatial resolution. However, the calculated spectral parameters are still not system independent due to difficulty in eliminating variations in the light distribution for different optical wavelengths. Changes in spectral parameters for the same absorber geometry but different optical illumination wavelengths needs to be carefully examined. A gelatin vessel phantom is used. The vessels contain red blood cells comprised of oxy, deoxy and methemoglobin induced using oxygen, sodium hydrosulfite and sodium nitrite, respectively. The samples were imaged using the VevoLAZR system at wavelengths 680 – 905 nm in steps of 15 nm. The radiofrequency (RF) signals were analyzed to calculate the spectral slope. The results were compared to simulated RF signals acquired using the mcxyz Monte Carlo package coupled to the solution of the PA wave equation using the Green’s function approach. Changes in the spectral slope as a function of optical wavelength were detected. For longer optical wavelengths, the spectral slope increased for deoxyhemoglobin, but decreased for oxyhemoglobin and methemoglobin. The changes in the spectral slope were correlated to changes in the fluence distribution as optical properties change for different wavelengths. The change in the spectral slope as a function of optical wavelength and chromophore content can potentially be used in spectral unmixing for better estimation of hemoglobin content.
Photoacoustic (PA) field calculations using a Green’s function approach is presented. The method has been applied to predict PA spectra generated by normal (discocyte) and pathological (stomatocyte) red blood cells (RBCs). The contours of normal and pathological RBCs were generated by employing a popular parametric model and accordingly, fitted with the Legendre polynomial expansions for surface parametrization. The first frequency minimum of theoretical PA spectrum approximately appears at 607 MHz for a discocyte and 410 MHz for a stomatocyte when computed from the direction of symmetry axis. The same feature occurs nearly at 247 and 331 MHz, respectively, for those particles when measured along the perpendicular direction. The average experimental spectrum for normal RBCs is found to be flat over a bandwidth of 150-500 MHz when measured along the direction of symmetry axis. For spherical RBCs, both the theoretical and experimental spectra demonstrate negative slope over a bandwidth of 250-500 MHz. Using the Green’s function method discussed, it may be possible to rapidly characterize cellular morphology from single-particle PA spectra.
The destruction of blood vessels is a commonly used cancer therapeutic strategy. Bleeding consequently follows and leads to the accumulation of blood in the interstitium. Photoacoustic (PA) imaging is well positioned to detect bleeding due to its sensitivity to hemoglobin. After treatment vascular disruption can occur within just a few hours, which leads to bleeding which might be detected using PA to assess therapeutic effectiveness. Deep micro-vessels cannot typically be resolved using acoustic-resolution PA. However, spectral analysis of PA signals may still permit assessment of bleeding. This paper introduces a theoretical model to simulate the PA signals from disrupted vessels using a fractal model. The fractal model uses bifurcated-cylinder bases to represent vascular trees. Vessels have circular absorption cross-sections. To mimic bleeding from blood vessels, the diffusion of hemoglobin from micro-vessels was simulated. The PA signals were computed and in the simulations were detected using a linear array transducer (30 MHz center frequency) for four different vascular trees (at 256 axial spatial locations/tree). The Fourier Transform of each beam-formed PA signal was computed and the power spectra were fitted to a straight line within the -6 dB bandwidth of the receiving transducer. When comparing the power spectra before and after simulated bleeding, the spectral slope and mid-band fit (MBF) parameters decreased by 0.12 dB/MHz and 2.12 dB, while the y-intercept did not change after 1 hour of simulated bleeding. The results suggest that spectral PA analysis is sensitive to changes in the concentration and spatial distribution of hemoglobin in tissue, and changes due to bleeding can be detected without the need to resolve individual vessels. The simulations support the applicability of PA imaging in cancer treatment monitoring by detecting micro-vessel disruption.
Many types of cancer therapies target the tumor microenvironment, causing biochemical and morphological changes in tissues. In therapies using ultrasound activated microbubbles, vascular collapse is typically reported. Red blood cells (RBCs) that leak out of the vasculature become exposed to the ceramide that is released from damaged endothelial cells. Ceramide can induce programmed cell death in RBCs (eryptosis), and is characterized by cell shrinkage, membrane blebbing and scrambling. Since the effect of eryptotic cells on generated photoacoustics (PA) signals has not been reported, we investigated the potential PA may have for cancer treatment monitoring by using PA spectral analysis to sense eryptosis. To induce eryptosis, C2-ceramide was added to RBC suspensions and that were incubated for 24 hours at 37°C. A control and ceramide-induced sample was imaged in a vessel phantom using a high frequency PA system (VevoLAZR, 10 – 45 MHz bandwidth) irradiated with multiple wavelengths ranging from 680 to 900 nm. PA spectral parameters were measured and linked to changes in RBCs as it underwent eryptosis. These samples were examined using optical microscopy, a blood gas analyzer and an integrating sphere setup to measure optical properties (wavelengths 600 – 900 nm). The results of the experiment demonstrate how PA spectral analysis can be used to identify eryptosis at a depth of more than 1 cm into the phantom using ultrasound derived the y-intercept and mid bandfit (MBF) parameters at optical wavelengths of 800 – 900 nm. These parameters were correlated to the morphological and biochemical changes that eryptotic RBCs display. The results establish the potential of PA in cancer treatment monitoring through sensing treatment induced eryptosis.
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.