Improving signal-to-noise ratio of a terahertz signal using a WaveNet-based neural network.

When acquiring a terahertz signal from a time-domain spectroscopy system, the signal is degraded by measurement noise and the information embedded in the signal is distorted. For high-performing terahertz applications, this study proposes a method for enhancing such a noise-degraded terahertz signal using machine learning that is applied to the raw signal after acquisition. The proposed method learns a function that maps the degraded signal to the clean signal using a WaveNet-based neural network that performs multiple layers of dilated convolutions. It also includes learnable pre- and post-processing modules that automatically transform the time domain where the enhancement process operates. When training the neural network, a data augmentation scheme is adopted to tackle the issue of insufficient training data. The comparative evaluation confirms that the proposed method outperforms other baseline neural networks in terms of signal-to-noise ratio. The proposed method also performs significantly better than the averaging of multiple signals, thereby facilitating the procurement of an enhanced signal without increasing the measurement time.

Machine Learning Techniques for THz Imaging and Time-Domain Spectroscopy.

Terahertz imaging and time-domain spectroscopy have been widely used to characterize the properties of test samples in various biomedical and engineering fields. Many of these tasks require the analysis of acquired terahertz signals to extract embedded information, which can be achieved using machine learning. Recently, machine learning techniques have developed rapidly, and many new learning models and learning algorithms have been investigated. Therefore, combined with state-of-the-art machine learning techniques, terahertz applications can be performed with high performance that cannot be achieved using modeling techniques that precede the machine learning era. In this review, we introduce the concept of machine learning and basic machine learning techniques and examine the methods for performance evaluation. We then summarize representative examples of terahertz imaging and time-domain spectroscopy that are conducted using machine learning.

Determining terahertz resonant peaks of biomolecules in aqueous environment.

The resonant peaks of biomolecules provide information on the molecules' physical and chemical properties. Although many biomolecules have resonant peaks in the terahertz region, it is difficult to observe their specific signals in aqueous environments. Hence, this paper proposes a method for determining these peaks. We found the specific resonant peaks of a modified nucleoside, 5-methlycytidine and modified HEK293T DNA in an aqueous solution through baseline correction. We evaluated the consistency of various fitting functions used for determining the peaks with various parameters. We separated two resonance peaks of 5-methlycytidine at 1.59 and 1.97 THz and for artificially methylated HEK293T DNA at 1.64 and 2.0 THz.

Highly Sensitive and Selective Detection of Steroid Hormones Using Terahertz Molecule-Specific Sensors.

Discrimination and quantification of trace amounts of steroid hormones in biological specimens are needed to elucidate their changing expression because their biological functions are responsible for the development and prevention of endocrine disorders. Although mass-spectrometry-based assays are most commonly recommended, development of a new type of highly sensitive and selective detection methods in clinical practices is needed. Here, we introduce a label-free type of terahertz molecule sensor capable of sensing and identifying progesterone and 17α-OH-progesterone selectively. Nanoslot-array-based sensing chips were used as launching pads for absorption cross-section enhancement of molecules at a reliable terahertz frequency. With use of nanoslots with resonances at 1.17 THz corresponding to intrinsic THz absorption resonance mode for progesterone and at 1.51 THz for 17α-OH-progesterone, respectively, each steroid shows prominent transmittance change in terms of its amount. In particular, the sensing performance has been much improved by controlling evaporation speed, in turn resulting in an efficient, homogeneous distribution of the molecules onto a sensing hot spot.

Unsaturated Drift Velocity of Monolayer Graphene.

We observe that carriers in graphene can be accelerated to the Fermi velocity without heating the lattice. At large Fermi energy | EF| > 110 meV, electrons excited by a high-power terahertz pulse ETHz relax by emitting optical phonons, resulting in heating of the graphene lattice and optical-phonon generation. This is owing to enhanced electron-phonon scattering at large Fermi energy, at which the large phase space is available for hot electrons. The emitted optical phonons cause carrier scattering, reducing the drift velocity or carrier mobility. However, for | EF| ≤ 110 meV, electron-phonon scattering rate is suppressed owing to the diminishing density of states near the Dirac point. Therefore, ETHz continues to accelerate carriers without them losing energy to optical phonons, allowing the carriers to travel at the Fermi velocity. The exotic carrier dynamics does not result from the massless nature, but the electron-optical-phonon scattering rate depends on Fermi level in the graphene. Our observations provide insight into the application of graphene for high-speed electronics without degrading carrier mobility.

Terahertz transmission control using polarization-independent metamaterials.

We present terahertz (THz) transmission control by several uniquely designed patterns of nano-slot antenna array. Collinearly aligned slot antenna arrays have been usually applied to THz filters with frequency band tunability by their geometry. Normally the amplitude in transmission (reflection) in the collinear alignment case can be varied via rotating the azimuthal angle with a sinusoidal trend, which can limit their utilization and performance only at fixed angle between the alignment of the resonant antennas and incident beam polarization. To pursue a variety of metamaterial uses, here, we present polarization-independent THz filters using variously aligned antenna array (asterisk, chlorophyll, and honeycomb patterns) in such counter-intuitive aspects. Besides, unprecedented multi resonance behaviors were observed in chlorophyll and honeycomb patterns, which can be explained with interferences by adjacent structures. The measured spectra were analyzed by harmonic oscillator model with simplified coupling between slots and their adjacent.

Enhancement of terahertz reflection tomographic imaging by interference cancellation between layers.

This paper proposes a method to enhance terahertz reflection tomographic imaging by interference cancellation between layers. When the gap between layers is small, the signal reflected on the upper layer interferes with that on the lower layer, which degrades the quality of the reconstructed tomographic image in the lower layer. The proposed method estimates the upper-layer reflection signal by system modeling, which is then eliminated from the acquired signal. In this way, it can provide the correct lower-layer reflection signal, thereby improving the quality of the lower-layer tomographic image. The performance of the proposed method was confirmed using computer simulation data and real terahertz reflection data.

Pre- and post-processing for tomographic reconstruction of terahertz time-domain spectroscopy.

Reflection-type terahertz tomography is obtained using time-domain spectroscopy. Due to different velocities of the terahertz ray in free space and inside a sample, the tomographic transverse plane is not obtained by a simple reconstruction using time index. A pre-processing method is proposed to compensate for the different velocities of the terahertz ray for tomographic reconstruction. Maximum intensity projection, averaging, and short-time Fourier transform are proposed as post-processing methods along the depth direction for the terahertz tomography. Log-scale display is also suggested for a better visualization. Some experimental results with the pre- and post-processing are demonstrated.

Measurement depth enhancement in terahertz imaging of biological tissues.

We demonstrate the use of a THz penetration-enhancing agent (THz-PEA) to enhance the terahertz (THz) wave penetration depth in tissues. The THz-PEA is a biocompatible material having absorption lower than that of water, and it is easily absorbed into tissues. When using glycerol as a THz-PEA, the peak value of the THz signal which was transmitted through the fresh tissue and reflected by a metal target, was almost doubled compared to that of tissue without glycerol. THz time-of-flight imaging (B-scan) was used to display the sequential glycerol delivery images. Enhancement of the penetration depth was confirmed after an artificial tumor was located below fresh skin. We thus concluded that the THz-PEA technique can potentially be employed to enhance the image contrast of the abnormal lesions below the skin.

Principle and applications of terahertz molecular imaging.

The principle, characteristics and applications of molecular imaging with terahertz electromagnetic waves are reviewed herein. The terahertz molecular imaging (TMI) technique uses nanoparticle probes to achieve dramatically enhanced sensitivity compared with that of conventional terahertz imaging. Surface plasmons, induced around the nanoparticles, raise the temperature of water in biological cells, and the temperature-dependent changes in the optical properties of water, which are large in the terahertz range, are measured differentially by terahertz waves. TMI has been applied to cancer diagnosis and nanoparticle drug delivery imaging. The technique is also compared with magnetic resonance imaging by using a dual-modality nanoparticle probe.

Gate-controlled nonlinear conductivity of Dirac fermion in graphene field-effect transistors measured by terahertz time-domain spectroscopy.

We present terahertz spectroscopic measurements of Dirac fermion dynamics from a large-scale graphene that was grown by chemical vapor deposition and on which carrier density was modulated by electrostatic and chemical doping. The measured frequency-dependent optical sheet conductivity of graphene shows electron-density-dependence characteristics, which can be understood by a simple Drude model. In a low carrier density regime, the optical sheet conductivity of graphene is constant regardless of the applied gate voltage, but in a high carrier density regime, it has nonlinear behavior with respect to the applied gate voltage. Chemical doping using viologen was found to be efficient in controlling the equilibrium Fermi level without sacrificing the unique carrier dynamics of graphene.

Epigenetic modification of gene expression in cancer cells by terahertz demethylation.

Terahertz (THz) radiation can affect the degree of DNA methylation, the spectral characteristics of which exist in the terahertz region. DNA methylation is an epigenetic modification in which a methyl (CH3) group is attached to cytosine, a nucleobase in human DNA. Appropriately controlled DNA methylation leads to proper regulation of gene expression. However, abnormal gene expression that departs from controlled genetic transcription through aberrant DNA methylation may occur in cancer or other diseases. In this study, we demonstrate the modification of gene expression in cells by THz demethylation using resonant THz radiation. Using an enzyme-linked immunosorbent assay, we observed changes in the degree of global DNA methylation in the SK-MEL-3 melanoma cell line under irradiation with 1.6-THz radiation with limited spectral bandwidth. Resonant THz radiation demethylated living melanoma cells by 19%, with no significant occurrence of apurinic/apyrimidinic sites, and the demethylation ratio was linearly proportional to the power of THz radiation. THz demethylation downregulates FOS, JUN, and CXCL8 genes, which are involved in cancer and apoptosis pathways. Our results show that THz demethylation has the potential to be a gene expression modifier with promising applications in cancer treatment.

Numerical Model for Cerebrovascular Hemodynamics with Indocyanine Green Fluorescence Videoangiography.

OBJECTIVE: The use of indocyanine green videoangiography (ICG-VA) to assess blood flow in the brain during cerebrovascular surgery has been increasing. Clinical studies on ICG-VA have predominantly focused on qualitative analysis. However, quantitative analysis numerical modelling for time profiling enables a more accurate evaluation of blood flow kinetics. In this study, we established a multiple exponential modified Gaussian (multi-EMG) model for quantitative ICG-VA to understand accurately the status of cerebral hemodynamics. METHODS: We obtained clinical data of cerebral blood flow acquired the quantitative analysis ICG-VA during cerebrovascular surgery. Varied asymmetric peak functions were compared to find the most matching function form with clinical data by using a nonlinear regression algorithm. To verify the result of the nonlinear regression, the mode function was applied to various types of data. RESULTS: The proposed multi-EMG model is well fitted to the clinical data. Because the primary parameters-growth and decay rates, and peak center and heights-of the model are characteristics of model function, they provide accurate reference values for assessing cerebral hemodynamics in various conditions. In addition, the primary parameters can be estimated on the curves with partially missed data. The accuracy of the model estimation was verified by a repeated curve fitting method using manipulation of missing data. CONCLUSION: The multi-EMG model can possibly serve as a universal model for cerebral hemodynamics in a comparison with other asymmetric peak functions. According to the results, the model can be helpful for clinical research assessment of cerebrovascular hemodynamics in a clinical setting.

Terahertz dynamic imaging of skin drug absorption.

Terahertz (THz) imaging is a nondestructive, label-free, rapid imaging technique which gives the possibility of a real-time tracing of drugs within the skin. We evaluated the feasibility of THz dynamic imaging for visualizing serial changes in the distribution and penetration of a topical agent, dimethyl sulfoxide (DMSO) containing ketoprofen, using excised mouse skins. THz imaging was performed for 6 h after drug application to the skin and was compared with the results obtained using the Franz cell diffusion test, a standard in vitro skin absorption test. THz dynamic reflection imaging showed that the reflection signals decreased rapidly during the early time period, and remained constant through the late time period. The area of drug permeation within the skin layer on THz imaging increased with time. The dynamic pattern of THz reflection signal decrease was similar to that of DMSO absorption analyzed by the Franz cell diffusion test, which indicates that THz imaging mainly reflects the DMSO component. This study demonstrates that THz imaging technique can be used for imaging the spatial distribution and penetration of drug-applied sites.

Fast terahertz reflection tomography using block-based compressed sensing.

In this paper, a new fast terahertz reflection tomography is proposed using block-based compressed sensing. Since measuring the time-domain signal on two-dimensional grid requires excessive time, reducing measurement time is highly demanding in terahertz tomography. The proposed technique directly reduces the number of sampling points in the spatial domain without modulation or transformation of the signal. Compressed sensing in spatial domain suggests a block-based reconstruction, which substantially reduces computational time without degrading the image quality. An overlap-average method is proposed to remove the block artifact in the block-based compressed sensing. Fast terahertz reflection tomography using the block-based compressed sensing is demonstrated with an integrated circuit and parched anchovy as examples.

Molecular imaging with terahertz waves.

We demonstrate a highly sensitive THz molecular imaging (TMI) technique involving differential modulation of surface plasmons induced on nanoparticles and obtain target specific in vivo images of cancers. This technique can detect quantities of gold nanoparticles as small as 15 µM in vivo. A comparison of TMI images with near infrared absorption images shows the superior sensitivity of TMI. Furthermore, the quantification property of TMI is excellent, being linearly proportional to the concentration of nanoparticles. The target specificity issue is also addressed at the ex vivo and cell levels. The high thermal sensitivity of TMI can help extend photonic-based photothermal molecular imaging researches from the in vitro level to the in vivo level. The TMI technique can be used for monitoring drug delivery processes and for early cancer diagnosis.

Transformation of terahertz vibrational modes of cytosine under hydration.

Cytosine and cytosine monohydrate are representative biomolecules for investigating the effect of hydrogen bonds in deoxyribonucleic acid. To better understand intermolecular interactions, such as hydrogen bonds, between nucleobases it is necessary to identify the low-frequency vibrational modes associated with intermolecular interactions and crystalline structures. In this study, we investigated the characteristic low-frequency vibrational modes of cytosine and cytosine monohydrate using terahertz time-domain spectroscopy (THz-TDS). The crystal geometry was obtained by the powder X-ray diffraction technique. The optimized atomic positions and the normal modes in the terahertz region were calculated using density functional theory (DFT), which agreed well with the experimental results. We found that overall terahertz absorption peaks of cytosine and cytosine monohydrate consist of collective vibrations mixed with intermolecular and intramolecular vibrations in mode character analysis, and that the most intense peaks of both samples involve remarkable intermolecular translational vibration. These results indicate that THz-TDS combined with DFT calculations including mode character analysis can be an effective method for understanding how water molecules contribute to the characteristics of the low-frequency vibrational modes by intermolecular vibrations with hydrogen bonding in biological and biomedical applications.

Smart drug-loaded polymer gold nanoshells for systemic and localized therapy of human epithelial cancer.

Near-infrared-light-sensitive multifunctional smart drug-loaded polymer gold nanoshells are fabricated as advanced prototypes, composed of chemotherapeutic agents (therapeutic antibody and anticancer drug-loaded polymeric nanoparticles) for systemic chemotherapy of human epithelial cancer and a polymer-based gold nanoshell for localized photothermal treatment by NIR light.

Nanoparticle-enabled terahertz imaging for cancer diagnosis.

This paper demonstrates the principle of the nanoparticle-contrast-agent-enabled terahertz imaging (CATHI) technique, which yields a dramatic sensitivity of the differential signal from cancer cells with nanoparticles. The terahertz (THz) reflection signal increased beam by 20% in the cancer cells with nanoparticles of gold nano-rods (GNRs) upon their irradiation with a infrared (IR) laser, due to the temperature rise of water in cancer cells by surface plasma ploritons. In the differential mode, the THz signal from the cancer cells with GNRs was 30 times higher than that from the cancer cells without GNRs. As the high sensitivity is achieved by the surface plasmon resonance through IR laser irradiation, the resolution of the CATHI technique can be as good as a few microns and THz endoscopy becomes more feasible.

A miniaturized fiber-coupled terahertz endoscope system.

In this study, we have designed, fabricated, and characterized a miniaturized optical fiber-coupled terahertz (THz) endoscope system. The endoscopic system utilized a photoconductive generator and detector driven by a mode-locked Ti:sapphire laser. In reflection mode, the endoscope showed a high signal-to-noise ratio and a wide frequency spectrum similar to the conventional THz time-domain spectroscopic system. The cross section of the endoscope including generator and detector head is (2 x 4 mm) x 6 mm, which is small enough to be inserted into a human body. For a feasibility test, the endoscopic system was used to measure reflective THz signals from the side wall of the mouth, tongue, and palm skin as well as from water for comparison. The absorption and refractive index of the side wall of the mouth and tongue were similar to those of water but those of the palm skin were less than water.