Microscale characterization of prostate biopsies tissues using optical coherence elastography and second harmonic generation imaging.

Photonics, especially optical coherence elastography (OCE) and second harmonic generation (SHG) imaging are novel high-resolution imaging modalities for characterization of biological tissues. Following our preliminary experience, we hypothesized that OCE and SHG imaging would delineate the microstructure of prostate tissue and aid in distinguishing cancer from the normal benign prostatic tissue. Furthermore, these approaches may assist in characterization of the grade of cancer, as well. In this study, we confirmed a high diagnostic accuracy of OCE and SHG imaging in the detection and characterization of prostate cancer for a large set of biopsy tissues obtained from men suspected to have prostate cancer using transrectal ultrasound (TRUS). The two techniques and methods described here are complementary, one depicts the stiffness of tissues and the other illustrates the orientation of collagen structure around the cancerous lesions. The results showed that stiffness of cancer tissue was ~57.63% higher than that of benign tissue (Young's modulus of 698.43±125.29 kPa for cancerous tissue vs 443.07±88.95 kPa for benign tissue with OCE. Using histology as a reference standard and 600 kPa as a cut-off threshold, the data analysis showed sensitivity and specificity of 89.6 and 99.8%, respectively. Corresponding positive and negative predictive values were 99.5 and 94.6%, respectively. There was a significant difference noticed in terms of Young's modulus for different Gleason scores estimated by OCE (P-value<0.05). For SHG, distinct patterns of collagen distribution were seen for different Gleason grade disease with computed quantification employing a ratio of anisotropic to isotropic (A:I ratio) and this correlated with disease aggressiveness.

Viscoelastic properties' characterization of corneal stromal models using non-contact surface acoustic wave optical coherence elastography (SAW-OCE).

Viscoelastic characterization of the tissue-engineered corneal stromal model is important for our understanding of the cell behaviors in the pathophysiologic altered corneal extracellular matrix (ECM). The effects of the interactions between stromal cells and different ECM characteristics on the viscoelastic properties during an 11-day culture period were explored. Collagen-based hydrogels seeded with keratocytes were used to replicate human corneal stroma. Keratocytes were seeded at 8 × 103 cells per hydrogel and with collagen concentrations of 3, 5 and 7 mg/ml. Air-pulse-based surface acoustic wave optical coherence elastography (SAW-OCE) was employed to monitor the changes in the hydrogels' dimensions and viscoelasticity over the culture period. The results showed the elastic modulus increased by 111%, 56% and 6%, and viscosity increased by 357%, 210% and 25% in the 3, 5 and 7 mg/ml hydrogels, respectively. To explain the SAW-OCE results, scanning electron microscope was also performed. The results confirmed the increase in elastic modulus and viscosity of the hydrogels, respectively, arose from increased fiber density and force-dependent unbinding of bonds between collagen fibers. This study reveals the influence of cell-matrix interactions on the viscoelastic properties of corneal stromal models and can provide quantitative guidance for mechanobiological investigations which require collagen ECM with tuneable viscoelastic properties.

Deep-learning approach for automated thickness measurement of epithelial tissue and scab using optical coherence tomography.

SIGNIFICANCE: In order to elucidate therapeutic treatment to accelerate wound healing, it is crucial to understand the process underlying skin wound healing, especially re-epithelialization. Epidermis and scab detection is of importance in the wound healing process as their thickness is a vital indicator to judge whether the re-epithelialization process is normal or not. Since optical coherence tomography (OCT) is a real-time and non-invasive imaging technique that can perform a cross-sectional evaluation of tissue microstructure, it is an ideal imaging modality to monitor the thickness change of epidermal and scab tissues during wound healing processes in micron-level resolution. Traditional segmentation on epidermal and scab regions was performed manually, which is time-consuming and impractical in real time. AIM: We aim to develop a deep-learning-based skin layer segmentation method for automated quantitative assessment of the thickness of in vivo epidermis and scab tissues during a time course of healing within a rodent model. APPROACH: Five convolution neural networks were trained using manually labeled epidermis and scab regions segmentation from 1000 OCT B-scan images (assisted by its corresponding angiographic information). The segmentation performance of five segmentation architectures was compared qualitatively and quantitatively for validation set. RESULTS: Our results show higher accuracy and higher speed of the calculated thickness compared with human experts. The U-Net architecture represents a better performance than other deep neural network architectures with 0.894 at F1-score, 0.875 at mean intersection over union, 0.933 at Dice similarity coefficient, and 18.28 µm at an average symmetric surface distance. Furthermore, our algorithm is able to provide abundant quantitative parameters of the wound based on its corresponding thickness maps in different healing phases. Among them, normalized epidermal thickness is recommended as an essential hallmark to describe the re-epithelialization process of the rodent model. CONCLUSIONS: The automatic segmentation and thickness measurements within different phases of wound healing data demonstrates that our pipeline provides a robust, quantitative, and accurate method for serving as a standard model for further research into effect of external pharmacological and physical factors.

Semisupervised representative learning for measuring epidermal thickness in human subjects in optical coherence tomography by leveraging datasets from rodent models.

SIGNIFICANCE: Morphological changes in the epidermis layer are critical for the diagnosis and assessment of various skin diseases. Due to its noninvasiveness, optical coherence tomography (OCT) is a good candidate for observing microstructural changes in skin. Convolutional neural network (CNN) has been successfully used for automated segmentation of the skin layers of OCT images to provide an objective evaluation of skin disorders. Such method is reliable, provided that a large amount of labeled data is available, which is very time-consuming and tedious. The scarcity of patient data also puts another layer of difficulty to make the model more generalizable. AIM: We developed a semisupervised representation learning method to provide data augmentations. APPROACH: We used rodent models to train neural networks for accurate segmentation of clinical data. RESULT: The learning quality is maintained with only one OCT labeled image per volume that is acquired from patients. Data augmentation introduces a semantically meaningful variance, allowing for better generalization. Our experiments demonstrate the proposed method can achieve accurate segmentation and thickness measurement of the epidermis. CONCLUSION: This is the first report of semisupervised representative learning applied to OCT images from clinical data by making full use of the data acquired from rodent models. The proposed method promises to aid in the clinical assessment and treatment planning of skin diseases.

Vitreous opacities in infants born full-term and preterm by handheld swept-source optical coherence tomography.

PURPOSE: To compare vitreous opacity density in infants born at term and in infants born prematurely using an investigational handheld swept-source optical coherence tomography (SS-OCT). METHODS: Infants born at term underwent imaging once between 12 and 48 hours after birth; infants born prematurely were imaged at each routine retinopathy of prematurity (ROP) examination. Three masked, trained graders analyzed images. Semiautomated methods were used to quantify vitreous opacity density, which was correlated with ROP severity based on indirect ophthalmoscopy, other SS-OCT findings, and medical comorbidities. RESULTS: Between April 2018 and June 2019, 251 SS-OCT imaging sessions were performed on 78 infants (49% female; 36% preterm, with mean birth weight of 1018 ± 338 g and gestational age of 28.6 ± 3.2 weeks). All SS-OCT sessions produced images of adequate quality. Punctate vitreous opacities were present in 25 of 28 term infants (89%) and 41 of 50 premature infants (82%). Dice coefficient and F1 scores for intergrader agreement were 0.99 ± 0.03 and 0.77 ± 0.31, respectively. Vitreous opacity density was 0.118 ± 0.187 in prematurely born infants and 0.031 ± 0.118 in infants born at term (P = 0.009). In the former, vitreous opacity density was associated with ROP zone (P = 0.044) and stage (P = 0.031), intraventricular hemorrhage (P = 0.028), subchorionic hemorrhage (P = 0.026), and African American race (P = 0.023). In the latter, vitreous opacity density was associated with maternal diabetes (P = 0.049). CONCLUSIONS: Our investigational handheld SS-OCT achieved high-quality vitreoretinal images. In our study cohort, punctate vitreous opacities were a frequent finding in infants born at term and those born prematurely, with increased density in those born prematurely, particularly those with severe ROP.

A Weighted Average Phase Velocity Inversion Model for Depth-Resolved Elasticity Evaluation in Human Skin In-Vivo.

OBJECTIVE: In current surface acoustic wave (SAW) elastography field, wavelength-depth inversion model is a straightforward and widely used inversion model for depth-resolved elasticity profile reconstruction. However, the elasticity directly evaluated from the wavelength-depth relationship is biased. Thus, a new inversion model, termed weighted average phase velocity (WAPV) inversion model, is proposed to provide depth-resolved Young's modulus estimate with better accuracy. METHODS: The forward model for SAW phase velocity dispersion curve generation was derived from the numerical simulations of SAWs in layered materials, and inversion was implemented by matching the measured phase velocity dispersion curve to the one generated from the forward model using the least squares fitting. Three two-layer agar phantoms with different top-layer thicknesses and one three-layer agar phantom were tested to validate the proposed inversion model. Then the model was demonstrated on human skin at various sites (palm, forearm and back of hand) in-vivo. RESULTS: In multi-layered agar phantoms, depth-resolved elasticity estimates provided by the model have a maximal total inversion error of 15.2% per sample after inversion error compensation. In in-vivo human skin, the quantified bulk Young's moduli (palm: 212 ± 78 kPa; forearm: 32 ± 11 kPa and back of hand: 29 ± 8 kPa) are comparable to the reference values in the literature. CONCLUSION: The WAPV inversion model can provide accurate depth-resolved Young's modulus estimates in layered biological soft tissues. SIGNIFICANCE: The proposed model can predict depth-resolved elasticity in layered biological soft tissues with a reasonable accuracy which traditional wavelength-depth inversion model cannot provide.

A novel automatic 3D stitching algorithm for optical coherence tomography angiography and its application in dermatology.

Optical coherence tomography (OCT) and OCT angiography (OCTA) techniques offer numerous advantages in clinical skin applications but the field of view (FOV) of current commercial systems are relatively limited to cover the entire skin lesion. The typical method to expand the FOV is to apply wide field objective lens. However, lateral resolution is often sacrificed when scanning with these lenses. To overcome this drawback, we developed an automated 3D stitching method for creating high-resolution skin structure and vascular volumes with large field of view, which was realized by montaging multiple adjacent OCT and OCTA volumes. The proposed stitching method is demonstrated by montaging 3 × 3 OCT and OCTA volumes (nine OCT/OCTA volumes as one data set with each volume covers 2.5 cm × 2.5 cm area) of healthy thin and thick skin from six volunteers. The proposed stitching protocol achieves high flexibility and repeatable for all the participants. Moreover, according to evaluation of structural similarity index and feature similarity index, our proposed stitched result has a superior similarity to single scanning protocol in large-scaled. We had also verified its improved performance through assessing metrics of vessel contrast-noise-ratio (CNR) from 2.07 ± 0.44 (single large-scaled scanning protocol) to 3.05 ± 0.51 (proposed 3 × 3 sub-volume stitching method).

Quantitative Handheld Swept-Source Optical Coherence Tomography Angiography in Awake Preterm and Full-Term Infants.

Purpose: To compare retinal vascular parameters acquired by handheld swept-source optical coherence tomography angiography (SS-OCTA) between nonsedated preterm and full-term infants. Methods: Preterm and full-term infants at the University of Washington Medical Center were enrolled. Retinal angiograms (nominal size ∼7 × 7 mm2) were obtained at each routine retinopathy of prematurity (ROP) screening session for preterms and once during the first 72 hours of life for full-terms. Macular vessel area density and nonperfusion area were evaluated on the binarized vasculature map in both small (1.5 × 1.5 mm) and large (3 × 3 mm) quadrants. Average vessel diameter and tortuosity values were obtained from each large vessel branch (length >200 µm). All vascular analyses used previously published algorithms. Results: Handheld SS-OCTA captured 31 of 55 (56%) high-quality volumes on 8 awake preterm infants (gestational age 28 ± 4 weeks, birth weight 891 ± 314 g, postmenstrual age at first imaging session 37 ± 2 weeks) and 48 of 54 (89%) volumes on 12 awake full-term infants (gestational age 39 ± 1 weeks, birth weight 3405 ± 329 g). Signal-to-noise ratio was 5.08 ± 1.52 dB in preterm and 4.90 ± 1.12 dB in full-term infants. Preterm infants had higher mean large vessel tortuosity compared to full-term infants (P = 0.004). The large nasal quadrant vessel area density of infants with stage 3 and/or pre-plus or worse ROP was higher than other preterm infants (P = 0.007). Conclusions: Although inadequate image quality limited usable imaging sessions, handheld SS-OCTA achieved adequate signal-to-noise ratio in nonsedated infants for quantitative retinal vascular parameter analysis. Translational Relevance: Large- and small-vessel parameters were associated with prematurity and ROP severity, respectively.

Feasibility study of using the dispersion of surface acoustic wave impulse for viscoelasticity characterization in tissue mimicking phantoms.

The aim of this study was to investigate the feasibility of utilizing the phase velocity dispersion of impulse surface acoustic wave (SAW) for viscoelasticity characterization of soft materials. The focused ultrasound transducer and the phase-sensitive optical coherence tomography were applied as the impulse SAW inducer and tracker, respectively. Three types of liquid-paraffin-based cream-in-agar phantoms were tested. Phase velocity dispersion curve was extracted using a Fourier transform-based phase velocity analysis algorithm. Viscoelastic parameters were obtained by fitting the dispersion curve of SAW into Rayleigh wave dispersion equation. The estimated viscoelasticity was compared with that from spherical indenter, ramp-hold relaxation testing for validation. Both results show an increasing trend in the elasticity and decreasing trend in the viscosity with the concentration of liquid-paraffin-based cream increasing in the samples. The proposed method has the capability of evaluating the viscoelastic properties of homogeneous soft tissue. By combining viscoelastic parameters estimated from the proposed method, the dispersive SAW-impulse-based viscosity-compensated elastography could be further developed.

Development of a clinical prototype of a miniature hand-held optical coherence tomography probe for prematurity and pediatric ophthalmic imaging.

We report a novel design and operation of a highly integrated miniature handheld OCT probe, with high-speed angiography function that can be used in clinical settings for young children and infants, providing rapid, non-invasive structural and angiographic imaging of the retina and choroid. The imaging system is operated at 200 kHz, with 3D OCT and OCTA scan time of 0.8 and 3.2 seconds, respectively, and the scanning angle on the pupil is ± 36°, covering the full perifoveal region. Operator assisting features of the direct-view iris camera and on-probe display are integrated into the hand-held probe, and the fixation target can display animations to attract the attention of young subjects. Compared to conventional OCT systems, the high-speed hand-held OCT system significantly improves the operator's experience and scanning efficiency, which is important for imaging infants. Imaging results indicate a significant reduction in total time consumption in pediatric ophthalmic imaging sessions, as well as the image quality of OCT angiography.

Visualizing choriocapillaris using swept-source optical coherence tomography angiography with various probe beam sizes.

Imaging choriocapillaris (CC) is a long-term challenge for commercial OCT angiography (OCTA) systems due to limited transverse resolution. Effects of transverse resolution on the visualization of a CC microvascular network are explored and demonstrated in this paper. We use three probe beams with sizes of ~1.12 mm, ~2.51 mm and ~3.50 mm at the pupil plane, which deliver an estimated transverse resolution at the retina of 17.5 µm, 8.8 µm and 7.0 µm, respectively, to investigate the ability of OCTA to resolve the CC capillary vessels. The complex optical microangiography algorithm is applied to extract blood flow in the CC slab. Mean retinal pigment epithelium (RPE) to CC (RPE-CC) distance, mean CC inter-vascular spacing and the magnitude in the radially-averaged power spectrum are quantified. We demonstrate that a clearer CC lobular capillary network is resolved in the angiograms provided by a larger beam size. The image contrast of the CC angiogram with a large beam size of 3.50 mm is 114% higher than that with a small beam size of 1.12 mm. While the measurements of the mean RPE-CC distance and CC inter-vascular spacing are almost consistent regardless of the beam sizes, they are more reliable and stable with the larger beam size of 3.50 mm. We conclude that the beam size is a key parameter for CC angiography if the purpose of the investigation is to visualize the individual CC capillaries.

Handheld swept-source optical coherence tomography with angiography in awake premature neonates.

BACKGROUND: Retinopathy of prematurity (ROP) can lead to retinal detachment and severe vision loss and is a common cause of childhood blindness. Optical coherence tomography angiography (OCTA) is a non-invasive imaging modality that can be used to detect potential abnormalities in the microvasculature in this population. The objective of this study is to assess the feasibility of a newly developed handheld swept source OCT (SS-OCT) device to successfully acquire structural vitreoretinal and retinal microvascular images in awake premature infants. METHODS: OCT and OCTA images were acquired at the time of routine ROP examinations from awake, unsedated preterm infants in the Neonatal Intensive Care Unit using a clinical research prototype handheld probe integrated with an SS-OCT system working at 1,060 nm wavelength and an imaging speed of 200,000 A-scans per second (200 kHz), enabling volume OCT and OCTA scans. Each volume was acquired with approximately 36˚ field of view (~6.3×6.3 mm in infants) in 4.8 s. Quality of acquired OCT and OCTA volume images, microvascular information, and vitreoretinal features were determined by 3-masked grader consensus. RESULTS: Twelve infants (5 females, mean gestational age 28.3 weeks, median birth weight 901 g, stages 0 to 3 ROP) underwent a total of 73 individual eye imaging sessions. High-quality OCT images of the fovea and the optic nerve were present in 69/73 (94.5%) and 56/73 (76.7%) scans, respectively. Vitreous bands were observed in 10/73 (13.7%); punctate hyperreflective vitreous opacities in 47/73 (64.4%); epiretinal membrane (ERM) in 6/73 (8.2%); and cystoid macular edema (CME) in 12/73 (16.4%) scans. Mild vessel elevation was noted in 3/73 (4.1%) images, and severe vessel elevation in 4/73 (5.5%) scans. OCTA images obtained in 8 awake infants revealed good quality images of the foveal microvasculature in 11/19 (58%) eye imaging sessions for 6/8 (75%) infants; and peripapillary microvasculature in 14/19 (74%) eye imaging sessions for 5/8 (63%) infants. CONCLUSIONS: The SS-OCTA handheld device can capture important vitreoretinal characteristics such as peripapillary and foveal microvasculature, as well as hyperreflective punctate vitreous opacities and tractional vitreous bands, which may predict ROP severity. These images were captured in awake, premature infants without the use of direct ocular contact, an eyelid speculum, or sedation.

Spatial resolution in dynamic optical coherence elastography.

Dynamic optical coherence elastography (OCE) tracks elastic wave propagation speed within tissue, enabling quantitative three-dimensional imaging of the elastic modulus. We show that propagating mechanical waves are mode converted at interfaces, creating a finite region on the order of an acoustic wavelength where there is not a simple one-to-one correspondence between wave speed and elastic modulus. Depending on the details of a boundary's geometry and elasticity contrast, highly complex propagating fields produced near the boundary can substantially affect both the spatial resolution and contrast of the elasticity image. We demonstrate boundary effects on Rayleigh waves incident on a vertical boundary between media of different shear moduli. Lateral resolution is defined by the width of the transition zone between two media and is the limit at which a physical inclusion can be detected with full contrast. We experimentally demonstrate results using a spectral-domain OCT system on tissue-mimicking phantoms, which are replicated using numerical simulations. It is shown that the spatial resolution in dynamic OCE is determined by the temporal and spatial characteristics (i.e., bandwidth and spatial pulse width) of the propagating mechanical wave. Thus, mechanical resolution in dynamic OCE inherently differs from the optical resolution of the OCT imaging system.

High-intensity-focused ultrasound and phase-sensitive optical coherence tomography for high resolution surface acoustic wave elastography.

Elastography has the ability of quantitatively evaluating the mechanical properties of soft tissue; thus it is helpful for diagnosis and treatment monitoring of many diseases, for example, skin diseases. Surface acoustic waves (SAWs) have been proven to be a non-invasive, non-destructive method for accurate characterization of tissue elastic properties. Current SAW elastography using high-energy laser pulse or mechanical shaker still have some problems. In order to improve SAW elastography in medical application, a new technique was proposed in this paper, which combines high-intensity-focused ultrasound as a SAWs impulse inducer and phase-sensitive optical coherence tomography as a SAWs detector. A 2% agar-agar phantom and ex-vivo porcine skin were tested. The data were processed by a new algorithm based on the Fourier analysis. The results show that the proposed method has the capability of quantifying the elastic properties of soft tissue-mimicking materials. The lateral resolution of the elastogram has been significantly improved and the different layers in heterogeneous material could also been distinguished. Our improved technique of SAW elastography has a large potential to be widely applied in clinical use for skin disease diagnosis and treatment monitoring.