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.
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.
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.
Pathological change tends to alter tissue mechanical properties, e.g. tissue stiffness. Current elastography technology use tissue stiffness as a signature to diagnose and localize diseases. Our team focus on vibrational optical coherence elastography (OCE) for its capability to increase signal to noise ratio as well as its high resolution comparing other elastography modalities. The result highly relies on the stimulation frequency for vibrational mode might change as frequency varies. A proper frequency range is required however, there hasn’t been a consensus among the research groups. In order to find the proper frequencies, several parameters measured from real experiment are input in transient model of ANSYS to simulate vibrational pattern of the sample with driving frequencies vary from 100Hz to 1000Hz. An upper limit of frequency has been discovered finally.
HIFU is a truly noninvasive, acoustic therapeutic technique that utilizes high intensity acoustic field in the focus to kill the targeted tissue for disease treatment purpose. The mechanical properties of targeted tissue changes before and after treatment, and this change can be accurately detected by shear wave elastography. Hence, shear wave elastography is usually used for monitoring HIFU treatment asynchronously. To improve the low spatial resolution in ultrasound shear wave elastography, and to perform diseases diagnosis, treatment and monitoring in the same system, a new setup that combines HIFU and PhS-OCT system was proposed in this study. This proposed setup could do 1) HIFU treatment when the transducer works at high energy level, 2) ultrasound induced shear wave optical coherence elastography for HIFU treatment asynchronous monitoring when the transducer works at low energy level. Ex-vivo bovine liver tissue was treated at the same energy level for different time (0s, 1s, 5s, 9s) in this research. Elastography was performed on the lesion area of the sample after HIFU treatment, and the elastogram was reconstructed by the time of flight time method. The elastogram results clearly show the boundary of HIFU lesion area and surrounding normal tissue, even for 1s treatment time. And the average elasticity of the lesion grows linearly as the treatment time increases. Combined with OCT needle probe, the proposed method has a large potential not only to be used for superficial diseases treatment, but also to be used for high-precision-demanded diseases treatment, e.g. nervous disease treatment.
Surface acoustic wave (SAW) elastography has been proven to be a non-invasive, non-destructive method for accurately characterizing tissue elastic properties. Current SAW elastography technique tracks generated surface acoustic wave impulse point by point which are a few millimeters away. Thus, reconstructed elastography has low lateral resolution. To improve the lateral resolution of current SAW elastography, a new method was proposed in this research. A M-B scan mode, high spatial resolution phase sensitive optical coherence tomography (PhS-OCT) system was employed to track the ultrasonically induced SAW impulse. Ex-vivo porcine skin specimen was tested using this proposed method. A 2D fast Fourier transform based algorithm was applied to process the acquired data for estimating the surface acoustic wave dispersion curve and its corresponding penetration depth. Then, the ex-vivo porcine skin elastogram was established by relating the surface acoustic wave dispersion curve and its corresponding penetration depth. The result from the proposed method shows higher lateral resolution than that from current SAW elastography technique, and the approximated skin elastogram could also distinguish the different layers in the skin specimen, i.e. epidermis, dermis and fat layer. This proposed SAW elastography technique may have a large potential to be widely applied in clinical use for skin disease diagnosis and treatment monitoring.
Ablative fractional skin laser is widely applied for various skin conditions, especially for cosmetic repairing and promoting the located drug delivery. Although the influence of laser treatment over the skin has been explored before in means of excision and biopsy with microscopy, these approaches are invasive, only morphological and capable of distorting the skin. In this paper the authors use fresh porcine skin samples irradiated by the lasers, followed by detected by using Optical Coherence Tomography (OCT). This advanced optical technique has the ability to present the high resolution structure image of treated sample. The results shows that laser beams can produce holes left on the surface after the irradiation. The depth of holes can be affected by changes of laser energy while the diameter of holes have no corresponding relation. Plus, OCT, as a valuable imaging technology, is capable of monitoring the clinical therapy procedure and assisting the calibration.
Prostate cancer (PCa) is a heterogeneous disease with multifocal origin. In current clinical care, the Gleason scoring system is the well-established diagnosis by microscopic evaluation of the tissue from trans-rectal ultrasound (TRUS) guided biopsies. Nevertheless, the sensitivity and specificity in detecting PCa can range from 40 to 50% for conventional TRUS B-mode imaging. Tissue elasticity is associated with the disease progression and elastography technique has recently shown promise in aiding PCa diagnosis. However, many cancer foci in the prostate gland has very small size less than 1 mm and those detected by medical elastography were larger than 2 mm. Hereby, we introduce optical coherence elastography (OCE) to quantify the prostate stiffness with high resolution in the magnitude of 10 µm. Following our feasibility study of 10 patients reported previously, we recruited 60 more patients undergoing 12-core TRUS guided biopsies for suspected PCa with a total of 720 biopsies. The stiffness of cancer tissue was approximately 57.63% higher than that of benign ones. Using histology as reference standard and cut-off threshold of 600kPa, the data analysis showed sensitivity and specificity of 89.6% and 99.8% respectively. The method also demonstrated potential in characterising different grades of PCa based on the change of tissue morphology and quantitative mechanical properties. In conclusion, quantitative OCE can be a reliable technique to identify PCa lesion and differentiate indolent from aggressive cancer.
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.