Deep neural network classification of in vivo burn injuries with different etiologies using terahertz time-domain spectral imaging

Research Advances in Terahertz Technology for Skin Detection

Background: With the continuous development of Terahertz technology and its high sensitivity to water, Terahertz technology has been widely applied in various research areas within the field of biomedicine, such as research onskin wounds and burns, demonstrating numerous advantages and potential. Objective: The aim of this study is to summarize and conclude the current research status of Terahertz radiation in skin wounds, burns, and melanoma. Additionally, it seeks toreveal the development status of Terahertz in skin wound models and analyze the short comings of Terahertz in detecting such models at the present stage. Methods: We retrieved relevant literature published from the inception of the Web of Science and CNKI databases up to 2024. The search terms included "THz," "Terahertz," "skin," "wound," "burn," and "melanoma." High-quality articles were included after rigorous screening. Results and Conclusions: This review explores the progress of terahertz radiation technology in the treatment and diagnosis of skin wounds and other related diseases. The results of its interaction with skin tissues provide valuable insights for future research. Terahertz radiation imaging has proven to be effective in assessing burn severity, capturing changes in edema, measuring exudates in dressings, assisting in burn grading and detection, and quantifying wound changes over time. Terahertz technology offers significant advantages in trauma assessment, which has accelerated its development and adoption in this field. (4) However, fs-THz radiation has been found to have the potential drawback of affecting wound healing. This finding necessitates careful consideration before application, and further research is warranted to explore its role in burn assessment and other medical applications.

THz Polarimetric Imaging of Carbon Fiber-Reinforced Composites Using the Portable Handled Spectral Reflection (PHASR) Scanner

Recent advancements in novel fiber-coupled and portable terahertz (THz) spectroscopic imaging technology have accelerated applications in nondestructive testing (NDT). Although the polarization information of THz waves can play a critical role in material characterization, there are few demonstrations of polarization-resolved THz imaging as an NDT modality due to the deficiency of such polarimetric imaging devices. In this paper, we have inspected industrial carbon fiber composites using a portable and handheld imaging scanner in which the THz polarizations of two orthogonal channels are simultaneously captured by two photoconductive antennas. We observed significant polarimetric differences between the two-channel images of the same sample and the resulting THz Stokes vectors, which are attributed to the anisotropic conductivity of carbon fiber composites. Using both polarimetric channels, we can visualize the superficial and underlying interfaces of the first laminate. These results pave the way for the future applications of THz polarimetry to the assessment of coatings or surface quality on carbon fiber-reinforced substrates.

A handheld polarimetric imaging device and calibration technique for accurate mapping of terahertz Stokes vectors

In recent years, handheld and portable terahertz instruments have been in rapid development for various applications ranging from non-destructive testing to biomedical imaging and sensing. For instance, we have deployed our Portable Handheld Spectral Reflection (PHASR) Scanners for in vivo full-spectroscopic imaging of skin burns in large animal models in operating room settings. In this paper, we debut the polarimetric version of the PHASR Scanner, and describe a generalized calibration technique to map the spatial and spectral dependence of the Jones matrix of an imaging scanner across its field of view. Our design is based on placement of two orthogonal photoconductive antenna (PCA) detectors separated by a polarizing beam splitter in the PHASR Scanner housing. We show that as few as three independent measurements of a well-characterized polarimetric calibration target are sufficient to determine the polarization state of the incident beam at the sample location, as well as to extract the Jones propagation matrix from the sample location to the detectors. We have tested the accuracy of our scanner by validating polarimetric measurements obtained from a birefringent crystal rotated to various angles, as compared to the theoretically predicted response of the sample. This new version of our PHASR scanner can be used for high-speed imaging and investigation of heterogeneity of polarization-sensitive samples in the field.

Review of machine learning for optical imaging of burn wound severity assessment

Significance

Over the past decade, machine learning (ML) algorithms have rapidly become much more widespread for numerous biomedical applications, including the diagnosis and categorization of disease and injury.

Aim

Here, we seek to characterize the recent growth of ML techniques that use imaging data to classify burn wound severity and report on the accuracies of different approaches.

Approach

To this end, we present a comprehensive literature review of preclinical and clinical studies using ML techniques to classify the severity of burn wounds.

Results

The majority of these reports used digital color photographs as input data to the classification algorithms, but recently there has been an increasing prevalence of the use of ML approaches using input data from more advanced optical imaging modalities (e.g., multispectral and hyperspectral imaging, optical coherence tomography), in addition to multimodal techniques. The classification accuracy of the different methods is reported; it typically ranges from ∼ 70 % to 90% relative to the current gold standard of clinical judgment.

Conclusions

The field would benefit from systematic analysis of the effects of different input data modalities, training/testing sets, and ML classifiers on the reported accuracy. Despite this current limitation, ML-based algorithms show significant promise for assisting in objectively classifying burn wound severity.

Design and characterization of a hyperbolic-elliptical lens pair in a rapid beam steering system for single-pixel terahertz spectral imaging of the cornea

Terahertz (THz) time-domain spectroscopy has been investigated for assessment of the hydration levels in the cornea, intraocular pressure, and changes in corneal topography. Previous efforts at THz imaging of the cornea have employed off-axis parabolic mirrors to achieve normal incidence along the spherical surface. However, this comes at the cost of an asymmetric field-of-view (FOV) and a long scan time because it requires raster-scanning of the collimated beam across the large mirror diameter. This paper proposes a solution by designing a pair of aspheric lenses that can provide a larger symmetric spherical FOV (9.6 mm) and reduce the scan time by two orders of magnitude using a novel beam-steering approach. A hyperbolic-elliptical lens was designed and optimized to achieve normal incidence and phase-front matching between the focused THz beam and the target curvature. The lenses were machined from a slab of high-density polyethylene and characterized in comparison to ray-tracing simulations by imaging several targets of similar sizes to the cornea. Our experimental results showed excellent agreement in the increased symmetric FOV and confirmed the reduction in scan time to about 3-4 seconds. In the future, this lens design process can be extended for imaging the sclera of the eye and other curved biological surfaces, such as the nose and fingers.

Triage of in vivo burn injuries and prediction of wound healing outcome using neural networks and modeling of the terahertz permittivity based on the double Debye dielectric parameters

The initial assessment of the depth of a burn injury during triage forms the basis for determination of the course of the clinical treatment plan. However, severe skin burns are highly dynamic and hard to predict. This results in a low accuracy rate of about 60 - 75% in the diagnosis of partial-thickness burns in the acute post-burn period. Terahertz time-domain spectroscopy (THz-TDS) has demonstrated a significant potential for non-invasive and timely estimation of the burn severity. Here, we describe a methodology for the measurement and numerical modeling of the dielectric permittivity of the in vivo porcine skin burns. We use the double Debye dielectric relaxation theory to model the permittivity of the burned tissue. We further investigate the origins of dielectric contrast between the burns of various severity, as determined histologically based on the percentage of the burned dermis, using the empirical Debye parameters. We demonstrate that the five parameters of the double Debye model can form an artificial neural network classification algorithm capable of automatic diagnosis of the severity of the burn injuries, and predicting its ultimate wound healing outcome by forecasting its re-epithelialization status in 28 days. Our results demonstrate that the Debye dielectric parameters provide a physics-based approach for the extraction of the biomedical diagnostic markers from the broadband THz pulses. This method can significantly boost dimensionality reduction of THz training data in artificial intelligence models and streamline machine learning algorithms.

Accurate and early prediction of the wound healing outcome of burn injuries using the wavelet Shannon entropy of terahertz time-domain waveforms

Significance

Severe burn injuries cause significant hypermetabolic alterations that are highly dynamic, hard to predict, and require acute and critical care. The clinical assessments of the severity of burn injuries are highly subjective and have consistently been reported to be inaccurate. Therefore, the utilization of other imaging modalities is crucial to reaching an objective and accurate burn assessment modality.

Aim

We describe a non-invasive technique using terahertz time-domain spectroscopy (THz-TDS) and the wavelet packet Shannon entropy to automatically estimate the burn depth and predict the wound healing outcome of thermal burn injuries.

Approach

We created 40 burn injuries of different severity grades in two porcine models using scald and contact methods of infliction. We used our THz portable handheld spectral reflection (PHASR) scanner to obtain the in vivo THz-TDS images. We used the energy to Shannon entropy ratio of the wavelet packet coefficients of the THz-TDS waveforms on day 0 to create supervised support vector machine (SVM) classification models. Histological assessments of the burn biopsies serve as the ground truth.

Results

We achieved an accuracy rate of 94.7% in predicting the wound healing outcome, as determined by histological measurement of the re-epithelialization rate on day 28 post-burn induction, using the THz-TDS measurements obtained on day 0. Furthermore, we report the accuracy rates of 89%, 87.1%, and 87.6% in automatic diagnosis of the superficial partial-thickness, deep partial-thickness, and full-thickness burns, respectively, using a multiclass SVM model.

Conclusions

The THz PHASR scanner promises a robust, high-speed, and accurate diagnostic modality to improve the clinical triage of burns and their management.

Terahertz PHASR Scanner with 2 kHz, 100 picosecond Time-Domain Trace Acquisition Rate and an Extended Field-of-View Based on a Heliostat Design

Recently, we introduced a Portable HAndheld Spectral Reflection (PHASR) Scanner to allow THz Time-Domain Spectroscopic (THz-TDS) imaging in clinical and industrial settings using a fiber-coupled and alignment-free telecentric beam steering design. The key limitations of the version 1.0 of the PHASR Scanner were its field-of-view and speed of time-domain trace acquisition. In this paper, we address these limitations by introducing a heliostat geometry for beam scanning to achieve an extended field-of-view, and by reconfiguring the Asynchronous OPtical Sampling (ASOPS) system to perform Electronically Controlled OPtical Sampling (ECOPS) measurements. The former change improved the deflection range of the beam, while also drastically reducing the coupling of the two scanning axes, the combination of which resulted in a larger than four-fold increase in the FOV area. The latter change significantly improves the acquisition speed and frequency domain performance simultaneously by improving measurement efficiency. To accomplish this, we characterized the non-linear time-axis sampling behavior of the electro-mechanical system in the ECOPS mode. We proposed methods to model and correct the non-linear time-axis distortions and tested the performance of the high-speed ECOPS trace acquisition. Therefore, here we introduce the PHASR Scanner version 2.0, which is capable of imaging a 40×27 mm2 FOV with 2000 traces per second over a 100 picosecond TDS range. This new scanner represents a significant leap towards translating the THz-TDS technology from the lab bench to the bedside for real-time clinical imaging applications.