Targeting Melanoma-Associated Fibroblasts (MAFs) with Activated γδ (Vδ2) T Cells: An In Vitro Cytotoxicity Model.

The tumor microenvironment (TME) has gained considerable scientific attention by playing a role in immunosuppression and tumorigenesis. Besides tumor cells, TME is composed of various other cell types, including cancer-associated fibroblasts (CAFs or MAFs when referring to melanoma-derived CAFs) and tumor-infiltrating lymphocytes (TILs), a subpopulation of which is labeled as γδ T cells. Since the current anti-cancer therapies using γδ T cells in various cancers have exhibited mixed treatment responses, to better understand the γδ T cell biology in melanoma, our research group aimed to investigate whether activated γδ T cells are capable of killing MAFs. To answer this question, we set up an in vitro platform using freshly isolated Vδ2-type γδ T cells and cultured MAFs that were biobanked from our melanoma patients. This study proved that the addition of zoledronic acid (1-2.5 µM) to the γδ T cells was necessary to drive MAFs into apoptosis. The MAF cytotoxicity of γδ T cells was further enhanced by using the stimulatory clone 20.1 of anti-BTN3A1 antibody but was reduced when anti-TCR γδ or anti-BTN2A1 antibodies were used. Since the administration of zoledronic acid is safe and tolerable in humans, our results provide further data for future clinical studies on the treatment of melanoma.

Preliminary Clinical Experience with a Novel Optical-Ultrasound Imaging Device on Various Skin Lesions.

A compact handheld skin ultrasound imaging device has been developed that uses co-registered optical and ultrasound imaging to provide diagnostic information about the full skin depth. The aim of the current work is to present the preliminary clinical results of this device. Using additional photographic, dermoscopic and ultrasonic images as reference, the images from the device were assessed in terms of the detectability of the main skin layer boundaries and characteristic image features. Combined optical-ultrasonic recordings of various types of skin lesions (melanoma, basal cell carcinoma, seborrheic keratosis, dermatofibroma, naevus, dermatitis and psoriasis) were taken with the device (N = 53) and compared with images captured with a reference portable skin ultrasound imager. The investigator and two additional independent experts performed the evaluation. The detectability of skin structures was over 90% for the epidermis, the dermis and the lesions. The morphological and echogenicity information observed for the different skin lesions were found consistent with those of the reference ultrasound device and relevant ultrasound images in the literature. The presented device was able to obtain simultaneous in-vivo optical and ultrasound images of various skin lesions. This has the potential for further investigations, including the preoperative planning of skin cancer treatment.

Automated Skin Lesion Classification on Ultrasound Images.

The growing incidence of skin cancer makes computer-aided diagnosis tools for this group of diseases increasingly important. The use of ultrasound has the potential to complement information from optical dermoscopy. The current work presents a fully automatic classification framework utilizing fully-automated (FA) segmentation and compares it with classification using two semi-automated (SA) segmentation methods. Ultrasound recordings were taken from a total of 310 lesions (70 melanoma, 130 basal cell carcinoma and 110 benign nevi). A support vector machine (SVM) model was trained on 62 features, with ten-fold cross-validation. Six classification tasks were considered, namely all the possible permutations of one class versus one or two remaining classes. The receiver operating characteristic (ROC) area under the curve (AUC) as well as the accuracy (ACC) were measured. The best classification was obtained for the classification of nevi from cancerous lesions (melanoma, basal cell carcinoma), with AUCs of over 90% and ACCs of over 85% obtained with all segmentation methods. Previous works have either not implemented FA ultrasound-based skin cancer classification (making diagnosis more lengthy and operator-dependent), or are unclear in their classification results. Furthermore, the current work is the first to assess the effect of implementing FA instead of SA classification, with FA classification never degrading performance (in terms of AUC or ACC) by more than 5%.

Automated seeding for ultrasound skin lesion segmentation.

The segmentation of cancer-suspicious skin lesions using ultrasound may help their differential diagnosis and treatment planning. Active contour models (ACM) require an initial seed, which when manually chosen may cause variations in segmentation accuracy. Fully-automated skin segmentation typically employs layer-by-layer segmentation using a combination of methods; however, such segmentation has not yet been applied on cancerous lesions. In the current work, fully automated segmentation is achieved in two steps: an automated seeding (AS) step using a layer-by-layer method followed by a growing step using an ACM. The method was tested on images of nevi, melanomas, and basal cell carcinomas from two ultrasound imaging systems (N=60), with all lesions being successfully located. For the seeding step, manual seeding (MS) was used as a reference. AS approached the accuracy of MS when the latter used an optimal bounding rectangle based on the ground truth (Sørensen-Dice coefficient (SDC) of 72.3 vs 74.6, respectively). The effect of varying the manual seed was also investigated; a 0.7 decrease in seed height and width caused a mean SDC of 54.6. The results show the robustness of automated seeding for skin lesion segmentation.

Enhancement of Acoustic Microscopy Lateral Resolution: A Comparison Between Deep Learning and Two Deconvolution Methods.

Scanning acoustic microscopy (SAM) provides high-resolution images of biological tissues. Since higher transducer frequencies limit penetration depth, image resolution enhancement techniques could help in maintaining sufficient lateral resolution without sacrificing penetration depth. Compared with existing SAM research, this work introduces two novelties. First, deep learning (DL) is used to improve lateral resolution of 180-MHz SAM images, comparing it with two deconvolution-based approaches. Second, 316-MHz images are used as ground truth in order to quantitatively evaluate image resolution enhancement. The samples used were mouse and rat brain sections. The results demonstrate that DL can closely approximate ground truth (NRMSE = 0.056 and PSNR = 28.4 dB) even with a relatively limited training set (four images, each smaller than 1 mm ×1 mm). This study suggests the high potential of using DL as a single image superresolution method in SAM.

A real-time data-based scan conversion method for single element ultrasound transducers.

The current work investigates the performance of a real-time scan conversion algorithm for generating a 2-D ultrasound image from a laterally scanned single-element ultrasound transducer, which has applications in point-of-care devices such as for skin imaging. The algorithm employs a fixed calibration curve to update a predefined image grid in real time. Simulations showed that the calibration curve (with a maximum of 1) is robust to changes in scatterer concentration (8.3×10-3 mean absolute error), signal to noise ratio (1.0×10-3 mean absolute error for -5 dB SNR), and can be accurately predicted from a small number (31) of point scatterers (6.9×10-3 mean absolute error). Good agreement was also found between the calibration curves obtained from simulated and experimental data (1.19×10-2 mean absolute error). The scan conversion algorithm was validated by evaluation of the position estimation errors on both simulations and experiments. Clinical images of skin lesions (N = 20) demonstrate the feasibility of the algorithm for real, non-homogeneous tissue. Use of a fixed calibration curve compared to an adaptive calibration curve gave similar accuracies in the scanning step size range of 150-350 µm (with an average overlap of the accuracy ranges of 92.94% for simulations and 42.83% for experiments), and a 350-fold improvement in computation time.

A Tensor Factorization Method for 3-D Super Resolution With Application to Dental CT.

Available super-resolution techniques for 3-D images are either computationally inefficient prior-knowledge-based iterative techniques or deep learning methods which require a large database of known low-resolution and high-resolution image pairs. A recently introduced tensor-factorization-based approach offers a fast solution without the use of known image pairs or strict prior assumptions. In this paper, this factorization framework is investigated for single image resolution enhancement with an offline estimate of the system point spread function. The technique is applied to 3-D cone beam computed tomography for dental image resolution enhancement. To demonstrate the efficiency of our method, it is compared to a recent state-of-the-art iterative technique using low-rank and total variation regularizations. In contrast to this comparative technique, the proposed reconstruction technique gives a 2-order-of-magnitude improvement in running time-2 min compared to 2 h for a dental volume of 282×266×392 voxels. Furthermore, it also offers slightly improved quantitative results (peak signal-to-noise ratio and segmentation quality). Another advantage of the presented technique is the low number of hyperparameters. As demonstrated in this paper, the framework is not sensitive to small changes in its parameters, proposing an ease of use.

Temperature dependence of speed of sound and attenuation of porcine left ventricular myocardium.

The temperature dependence of soft tissue acoustic properties is relevant for monitoring tissue hyperthermia. In the current work, the propagation speed and attenuation of healthy porcine left ventricular myocardium (N=5) was investigated in a frequency range relevant for clinical diagnostic imaging, i.e. 2.5-13.0MHz. Each tissue sample was held in a water bath at a temperature T=25°C, heated to 45°C, and allowed to cool back down to 25°C. Due to initial tissue swelling, the data for decreasing temperatures was considered more reliable. In this case, the slope of the phase velocity versus temperature relation was measured to be 1.10±0.04m/s/°C, and the slope of the attenuation was -0.11±0.04dB/cm/°C at 10MHz, or -0.0041±0.0015dB/cm/MHz1.4336/°C as a function of frequency.

Comparison of Two Inexpensive Rapid Prototyping Methods for Manufacturing Filament Target Ultrasound Phantoms.

Current use of 3-D printers to manufacture ultrasound phantoms is limited to relatively expensive photopolymer jetting printers. The present work investigates the feasibility of using two common and inexpensive 3-D printer technologies, fused deposition modeling (FDM) and digital light processing (DLP), to print custom filament target phantoms. Acoustic characteristics obtained from printed solid blocks indicated that the printing materials-acrylonitrile butadiene styrene and polylactic acid for FDM and a photopolymer for DLP printing-were appropriate for use as scatterers. A regular grid of filaments was printed to study printing accuracy. As a proof of concept of the phantom manufacturing process, a complex pattern of filament targets was placed in de-ionized water to create a phantom, which was then imaged using an ultrasound imager. The pattern was clearly identifiable, although multiple reflections were observed, which underscores the importance of future work to enhance printing resolution. This goal is deemed possible using improvement of the DLP printing setup.

Temperature increase induced by modulated electrohyperthermia (oncothermia®) in the anesthetized pig liver.

AIM OF STUDY: Is to show the intrahepatic temperature development in anesthetized pig. MATERIALS AND METHODS: Temperature development in the liver of anesthetized pig is measured to study the thermal effects of capacitive coupled energy transfer. The treatment was made by modulated electrohyperthermia (mEHT, trade name: oncothermia ®), controlled by a fluoroptical temperature sensing positioned by the ultrasound-guided process. Various fits of coupling were studied. RESULTS: The intrahepatic temperature at the end of the treatment ranged 40.5-44.8°C, while the skin temperature ranged 36.8-41.8°C depending on the coupling arrangement. CONCLUSION: mEHT is a feasible method to deliver deep heat to the liver of an anesthetized pig.

Experimental validation of a convolution- based ultrasound image formation model using a planar arrangement of micrometer-scale scatterers.

The shift-invariant convolution model of ultrasound is widely used in the literature, for instance to generate fast simulations of ultrasound images. However, comparison of the resulting simulations with experiments is either qualitative or based on aggregate descriptors such as envelope statistics or spectral components. In the current work, a planar arrangement of 49-µm polystyrene microspheres was imaged using macrophotography and a 4.7-MHz ultrasound linear array. The macrophotograph allowed estimation of the scattering function (SF) necessary for simulations. Using the coefficient of determination R(2) between real and simulated ultrasound images, different estimates of the SF and point spread function (PSF) were tested. All estimates of the SF performed similarly, whereas the best estimate of the PSF was obtained by Hanningwindowing the deconvolution of the real ultrasound image with the SF: this yielded R(2) = 0.43 for the raw simulated image and R(2) = 0.65 for the envelope-detected ultrasound image. R(2) was highly dependent on microsphere concentration, with values of up to 0.99 for regions with scatterers. The results validate the use of the shift-invariant convolution model for the realistic simulation of ultrasound images. However, care needs to be taken in experiments to reduce the relative effects of other sources of scattering such as from multiple reflections, either by increasing the concentration of imaged scatterers or by more careful experimental design.

Passive acoustic mapping utilizing optimal beamforming in ultrasound therapy monitoring.

Passive acoustic mapping (PAM) is a promising imaging method that enables real-time three-dimensional monitoring of ultrasound therapy through the reconstruction of acoustic emissions passively received on an array of ultrasonic sensors. A passive beamforming method is presented that provides greatly improved spatial accuracy over the conventionally used time exposure acoustics (TEA) PAM reconstruction algorithm. Both the Capon beamformer and the robust Capon beamformer (RCB) for PAM are suggested as methods to reduce interference artifacts and improve resolution, which has been one of the experimental issues previously observed with TEA. Simulation results that replicate the experimental artifacts are shown to suggest that bubble interactions are the chief cause. Analysis is provided to show that these multiple bubble artifacts are generally not reduced by TEA, while Capon-based methods are able to reduce the artifacts. This is followed by experimental results from in vitro experiments and in vivo oncolytic viral therapy trials that show improved results in PAM, where RCB is able to more accurately localize the acoustic activity than TEA.

Scattering by single physically large and weak scatterers in the beam of a single-element transducer.

Quantitative ultrasound techniques are generally applied to characterize media whose scattering sites are considered to be small compared to a wavelength. In this study, the backscattered response of single weakly scattering spheres and cylinders with diameters comparable to the beam width of a 2.25 MHz single-element transducer were simulated and measured in the transducer focal plane to investigate the impact of physically large scatterers. The responses from large single spherical scatterers at the focus were found to closely match the plane-wave response. The responses from large cylindrical scatterers at the focus were found to differ from the plane-wave response by a factor of f(-1). Normalized spectra from simulations and measurements were in close agreement: the fall-off of the responses as a function of lateral position agreed to within 2 dB for spherical scatterers and to within 3.5 dB for cylindrical scatterers. In both measurement and simulation, single scatterer diameter estimates were biased by less than 3% for a more highly focused transducer compared to estimates for a more weakly focused transducer. The results suggest that quantitative ultrasound techniques may produce physically meaningful size estimates for media whose response is dominated by scatterers comparable in size to the transducer beam.

Cavitation-enhanced delivery of insulin in agar and porcine models of human skin.

Ultrasound-assisted transdermal insulin delivery offers a less painful and less invasive alternative to subcutaneous insulin injections. However, ultrasound-based drug delivery, otherwise known as sonophoresis, is a highly variable phenomenon, in part dependent on cavitation. The aim of the current work is to investigate the role of cavitation in transdermal insulin delivery. Fluorescently stained, soluble Actrapid insulin was placed on the surface of human skin-mimicking materials subjected to 265 kHz, 10% duty cycle focused ultrasound. A confocally and coaxially aligned 5 MHz broadband ultrasound transducer was used to detect cavitation. Two different skin models were used. The first model, 3% agar hydrogel, was insonated with a range of pressures (0.25-1.40 MPa peak rarefactional focal pressure-PRFP), with and without cavitation nuclei embedded within the agar at a concentration of 0.05% w/v. The second, porcine skin was insonated at 1.00 and 1.40 MPa PRFP. In both models, fluorescence measurements were used to determine penetration depth and concentration of delivered insulin. Results show that in agar gel, both insulin penetration depth and concentration only increased significantly in the presence of inertial cavitation, with up to a 40% enhancement. In porcine skin the amount of fluorescent insulin was higher in the epidermis of those samples that were exposed to ultrasound compared to the control samples, but there was no significant increase in penetration distance. The results underline the importance of instigating and monitoring inertial cavitation during transdermal insulin delivery.

Variation of ultrasound image lateral spectrum with assumed speed of sound and true scatterer density.

One method of estimating sound speed in diagnostic ultrasound imaging consists of choosing the speed of sound that generates the sharpest image, as evaluated by the lateral frequency spectrum of the squared B-mode image. In the current work, simulated and experimental data on a typical (47 mm aperture, 3.3-10.0 MHz response) linear array transducer are used to investigate the accuracy of this method. A range of candidate speeds of sound (1240-1740 m/s) was used, with a true speed of sound of 1490 m/s in simulations and 1488 m/s in experiments. Simulations of single point scatterers and two interfering point scatterers at various locations with respect to each other gave estimate errors of 0.0-2.0%. Simulations and experiments of scatterer distributions with a mean scatterer spacing of at least 0.5 mm gave estimate errors of 0.1-4.0%. In the case of lower scatterer spacing, the speed of sound estimates become unreliable due to a decrease in contrast of the sharpness measure between different candidate speeds of sound. This suggests that in estimating speed of sound in tissue, the region of interest should be dominated by a few, sparsely spaced scatterers. Conversely, the decreasing sensitivity of the sharpness measure to speed of sound errors for higher scatterer concentrations suggests a potential method for estimating mean scatterer spacing.

Histology-based simulations of ultrasound imaging: methodology.

Simulations of ultrasound (US) images based on histology may shed light on the process by which microscopic tissue features translate to a US image and may enable predictions of feature detectability as a function of US system parameters. This technical note describes how whole-slide hematoxylin and eosin-stained histology images can be used to generate maps of fractional change in bulk modulus, whose convolution with the impulse response of the US system yields simulated US images. The method is illustrated by two canine mastocytoma histology images, one with and the other without signs of intra-operative hemorrhaging. Quantitative comparisons of the envelope statistics with corresponding clinical US images provide preliminary validation of the method.

Spatiotemporal monitoring of high-intensity focused ultrasound therapy with passive acoustic mapping.

PURPOSE: To demonstrate feasibility of monitoring high-intensity focused ultrasound (HIFU) treatment with passive acoustic mapping of broadband and harmonic emissions reconstructed from filtered-channel radiofrequency data in ex vivo bovine tissue. MATERIALS AND METHODS: Both passive acoustic emissions and B-mode images were recorded with a diagnostic ultrasound machine during 180 HIFU exposures of five freshly excised, degassed bovine livers. Tissue was exposed to peak rarefactional pressures between 3.6 and 8.0 MPa for 2, 5, or 10 seconds. The B-mode images were analyzed for hyperechoic activity, and threshold levels were determined for the harmonic (1.17 mJ) and broadband (0.0137 mJ) components of the passively reconstructed source energy to predict tissue ablation. Both imaging methods were compared with tissue lesions after exposure to determine their spatial accuracy and their capability to help predict presence of ablated tissue. Performance of both methods as detectors was compared (matched-pair test design). RESULTS: Passive mapping successfully aided prediction of the presence of tissue ablation more often than did conventional hyperechoic images (49 of 58 [84%] vs 31 of 58 [53%], P < .001). At 5.4-6.3-MPa exposures, sensitivity, specificity, negative predictive value, and positive predictive value of the two methods, respectively, were 15 of 20 versus five of 21 (P = .006), eight of nine versus eight of nine (P = .72), 15 of 16 versus five of six (P = .53), and eight of 13 versus eight of 24 (P = .011). Across HIFU exposure amplitude ranges, passive acoustic mapping also aided correct prediction of the visually detected location of ablation following tissue sectioning in 42 of 45 exposures for which the harmonic and broadband threshold levels for tissue ablation were exceeded. Early cavitation activity indicated the focal position within the tissue before irreversible tissue damage occurred. CONCLUSION: Passive acoustic mapping significantly outperformed the conventional hyperecho technique as an ultrasound-based HIFU monitoring method, as both a detector of lesion occurrence and a method of mapping the position of ablated tissue.

Passive cavitation mapping with temporal sparsity constraint.

The spatial resolution of cavitation maps generated from passive recordings of cavitation emissions is compromised by the bandlimited nature of the recordings. Deconvolution based on the assumption that cavitation consists of a sparse series of discrete events allows the recovery of frequency components that are not only outside the frequency band of the receivers, but may also have been attenuated by the medium before being detectable. In the current work, two sparse deconvolution techniques, matching pursuit and basis pursuit, were applied to simulated and experimental cavitation recordings before they were beamformed to provide passive maps of cavitation activity. Matching pursuit was shown to reduce the maximal diameter of the point spread function by almost a third, at the cost of greater susceptibility to inter-source interference. In contrast, although basis pursuit causes an almost 20% increase in the maximal diameter of the point spread function, its application to experimental data appears to enhance the ability of passive mapping to resolve multiple sources.

Passive cavitation mapping for localization and tracking of bubble dynamics.

Current acoustic techniques for studying cavitation dynamics are only readily applicable to single-bubble activity, while optical methods can only be used in transparent media. However, multi-bubble cavitation often occurs in opaque media such as biological tissue. Here, the signals received passively by each of the 64 channels of a diagnostic ultrasound array are used to localize and separate emissions from several bubble clusters cavitating in agar gel, thereby providing a method of observing cavitation dynamics. The method has a high spatiotemporal resolution and is applicable to cavitation in opaque media.

Passive spatial mapping of inertial cavitation during HIFU exposure.

A novel method for mapping inertial cavitation activity during high-intensity focused ultrasound (HIFU) exposure is presented. Inertial cavitation has been previously shown to result in increased heat deposition and to be associated with broadband noise emissions that can be readily monitored using a passive receiver without interference from the main HIFU signal. In the present study, the signals received passively by each of 64 elements on a standard diagnostic array placed coaxially with the HIFU transducer are combined using time exposure acoustics to generate maps of inertially cavitating regions during HIFU exposure of an agar-based tissue-mimicking material. The technique is shown to be effective in localizing single-bubble activity, as well as contiguous and disjoint cavitating regions instigated by creating regions of lower cavitation threshold within the tissue phantom. The cavitation maps obtained experimentally are also found to be in good agreement with computational simulations and theoretical predictions. Unlike B-mode imaging, which requires interleaving with the HIFU pulse, passive array-based mapping of cavitation activity is possible during HIFU exposure. If cavitating regions can be directly correlated to increased tissue damage, this novel cavitation mapping technique could enable real-time HIFU treatment monitoring.