Terahertz spectroscopic diagnosis of early blast-induced traumatic brain injury in rats

A new horizon for neuroscience: terahertz biotechnology in brain research

Terahertz biotechnology has been increasingly applied in various biomedical fields and has especially shown great potential for application in brain sciences. In this article, we review the development of terahertz biotechnology and its applications in the field of neuropsychiatry. Available evidence indicates promising prospects for the use of terahertz spectroscopy and terahertz imaging techniques in the diagnosis of amyloid disease, cerebrovascular disease, glioma, psychiatric disease, traumatic brain injury, and myelin deficit. In vitro and animal experiments have also demonstrated the potential therapeutic value of terahertz technology in some neuropsychiatric diseases. Although the precise underlying mechanism of the interactions between terahertz electromagnetic waves and the biosystem is not yet fully understood, the research progress in this field shows great potential for biomedical noninvasive diagnostic and therapeutic applications. However, the biosafety of terahertz radiation requires further exploration regarding its two-sided efficacy in practical applications. This review demonstrates that terahertz biotechnology has the potential to be a promising method in the field of neuropsychiatry based on its unique advantages.

High-sensitivity THz-ATR imaging of cerebral ischemia in a rat model

The fast label-free detection of the extent and degree of cerebral ischemia has been the difficulty and hotspot for precise and accurate neurosurgery. We experimentally demonstrated that the fresh cerebral tissues at different ischemic stages within 24 hours can be well distinguished from the normal tissues using terahertz (THz) attenuated total reflection (ATR) imaging system. It was indicated that the total reflectivity of THz wave for ischemic cerebral tissues was lower than that for normal tissues. Especially, compared to the images stained with 2,3,5-triphenyl tetrazolium chloride (TTC), the ischemic tissues can be detected using THz wave with high sensitivity as early as the ischemic time of 2.5 hours, where THz images showed the ischemic areas became larger and diffused as the ischemic time increasing. Furthermore, the THz spectroscopy of cerebral ischemic tissues at different ischemic times was obtained in the range of 0.5-2.0 THz. The absorption coefficient of ischemic tissue increased with the increase of ischemic time, whereas the refractive index decreased with prolonging the ischemic time. Additionally, it was found from hematoxylin and eosin (H&E) staining microscopic images that, with the ischemic time increasing, the cell size and cell density of the ischemic tissues decreased, whereas the intercellular substance of the ischemic tissues increased. The result showed that THz recognition mechanism of the ischemia is mainly based on the increase of intercellular substance, especially water content, which has a stronger impact on absorption of THz wave than that of cell density. Thus, THz imaging has great potential for recognition of cerebral ischemia and it may become a new method for intraoperative real-time guidance, recognition in situ, and precise excision.

Physics-assisted machine learning for THz time-domain spectroscopy: sensing leaf wetness

Signal processing techniques are of vital importance to bring THz spectroscopy to a maturity level to reach practical applications. In this work, we illustrate the use of machine learning techniques for THz time-domain spectroscopy assisted by domain knowledge based on light-matter interactions. We aim at the potential agriculture application to determine the amount of free water on plant leaves, so-called leaf wetness. This quantity is important for understanding and predicting plant diseases that need leaf wetness for disease development. The overall transmission of 12,000 distinct water droplet patterns on a plastized leaf was experimentally acquired using THz time-domain spectroscopy. We report on key insights of applying decision trees and convolutional neural networks to the data using physics-motivated choices. Eventually, we discuss the generalizability of these models to determine leaf wetness after testing them on cases with increasing deviations from the training set.

Study of low-frequency spectroscopic characteristics of γ-aminobutyric acid with THz and low-wavenumber Raman spectroscopy

γ-aminobutyric (GABA) is the most important inhibitory neurotransmitier in vertebrate central nervous systems. The content level of GABA is related to the different degree of malignancy gliomas. Thus, it can be considered a promising glioma biomarker. In this paper, the spectroscopic properties of GABA have been characterized by combining the THz spectroscopy with low-wavenumber Raman spectroscopy. The experimental results showed that, GABA exhibited three absorption peaks and three refractive index peaks in the range of 0.6-2.1 THz. The limit of detection can reach up to 0.428 % based on the absorption coefficient at the peak of 2.04 THz. Moreover, the low-wavenumber Raman spectrum of GABA showed seven characteristic peaks at 41.0, 50.8, 58.8, 77.2, 98.8, 115.6, 141.2 cm-1 in 0-150 cm-1 region. Moreover, the THz and low-wavenumber theoretical spectra of GABA were simulated with solid-state density function theory, respectively. The calculated results were in good agreement with the experimental observations. On the basis of calculated result, the vibrational motions of each THz and Raman characteristic modes were quantitatively decomposed by analytical mode-decoupling method, where the contribution percentages of external translation, external librations and intramolecular vibration of each vibration modes were analyzed Furthermore, the low-frequency characteristics of GABA was analyzed by combining the THz and low-wavenumber Raman spectroscopy. It is beneficial for the structural information analysis and quantitative identification of biomarker GABA in early stage diagnosis of glioma.

Raman spectroscopic diagnosis of blast-induced traumatic brain injury in rats combined with machine learning

Blast-induced traumatic brain injury (bTBI) is a kind of nervous system disease, which results in a major health and economic problem to society. However, the rapid and label-free detection method with high sensitivity is still in great demand for the diagnosis of bTBI, especially for mild bTBI. In this paper, we report a new strategy for bTBI diagnosis through hippocampus and hypothalamus tissues based on Raman spectroscopy. The spectral characteristics of hippocampus and hypothalamus tissues of experimental bTBI in rats have been investigated for mild and moderate degrees at 3 h, 6 h, 24 h, 48 h, 72 h after blast exposure. The results show that the Raman spectra of mild and moderate bTBIs in 300-1700 cm-1 and 2800-3000 cm-1 regions exhibit significant differences at different time points compared with the control group. The main reason is the content change of proteins and lipids in hippocampus and hypothalamus tissues after bTBI. Moreover, four machine learning algorithms are used to automatically identify mild and moderate bTBIs at different time points (a total of 11 groups). The highest diagnostic accuracies are up to 95.3% and 88.5% based on Raman spectra of hippocampus and hypothalamus tissue, respectively. In addition, the classification performance of linear discriminant analysis classifier has been improved after data fusion. It is suggested that there has great potential as an alternative method for high-sensitive, rapid, label-free, economical diagnosis of bTBI.

High resolution terahertz ATR frequency-domain spectroscopy for monitoring spinal cord injury in rats

Traumatic spinal cord injury (SCI) can lead to permanent neurological impairment, underscoring the urgency of regular therapeutic intervention and monitoring. In this study, we propose a new strategy for monitoring spinal cord injury through serum based on high-resolution THz attenuated total reflection frequency domain spectroscopy (THz-ATR-FDS). We demonstrated serum spectral differences at different time points after experimental SCI in rats. We also studied the relationship between serum lipid concentration and the time of SCI, which revealed the potential of lipid molecules as biomarkers of SCI. In addition, based on the principal component analysis (PCA) and least squares regression (LSR) models, the quantitative relationship between the refractive index spectrum and lipid concentration in serum was automatically analyzed. This work highlights terahertz spectroscopy as a promising tool for label-free, periodic, and efficient monitoring of SCI.

Biomedical application of terahertz imaging technology: a narrative review

Background and Objective

Terahertz (THz) imaging has wide applications in biomedical research due to its properties, such as non-ionizing, non-invasive and distinctive spectral fingerprints. Over the past 6 years, the application of THz imaging in tumor tissue has made encouraging progress. However, due to the strong absorption of THz by water, the large size, high cost, and low sensitivity of THz devices, it is still difficult to be widely used in clinical practice. This paper provides ideas for researchers and promotes the development of THz imaging in clinical research.

Methods

The literature search was conducted in the Web of Science and PubMed databases using the keywords "Terahertz imaging", "Breast", "Brain", "Skin" and "Cancer". A total of 94 English language articles from 1 January, 2017 to 30 December, 2022 were reviewed.

Key Content and Findings

In this review, we briefly introduced the recent advances in THz near-field imaging, single-pixel imaging and real-time imaging, the applications of THz imaging for detecting breast, brain and skin tissues in the last 6 years were reviewed, and the advantages and existing challenges were identified. It is necessary to combine machine learning and metamaterials to develop real-time THz devices with small size, low cost and high sensitivity that can be widely used in clinical practice. More powerful THz detectors can be developed by combining graphene, designing structures and other methods to improve the sensitivity of the devices and obtain more accurate information. Establishing a THz database is one of the important methods to improve the repeatability and accuracy of imaging results.

Conclusions

THz technology is an effective method for tumor imaging. We believe that with the joint efforts of researchers and clinicians, accurate, real-time, and safe THz imaging will be widely applied in clinical practice in the future.

Terahertz Irradiation Improves Cognitive Impairments and Attenuates Alzheimer's Neuropathology in the APPSWE/PS1DE9 Mouse: A Novel Therapeutic Intervention for Alzheimer's Disease

Alzheimer's disease (AD) is a progressive neurodegenerative disease characterized by the deposition of amyloid-β (Aβ), neurofibrillary tangles, neuroinflammation, and neurodegeneration in the brain. In recent years, considering the unsatisfied benefits of pharmacological therapies, non-pharmacological therapy has become a research hotspot for AD intervention. Terahertz (THz) waves with a range between microwave and infrared regions in the electromagnetic spectrum and high permeability to a wide range of materials have great potential in the bioengineering field. However, its biological impacts on the central nervous system, under either physiological or pathological conditions, are poorly investigated. In this study, we first measured the 0.14 THz waves penetration across the skull of a C57BL/6 mouse and found the percentage of THz penetration to be ~70%, guaranteeing that THz waves can reach the relevant brain regions. We then exposed the APPSWE/PS1DE9 mouse model of AD to repeated low-frequency THz waves on the head. We demonstrated that THz waves treatment significantly improved the cognitive impairment and alleviated AD neuropathology including Aβ deposition and tau hyperphosphorylation in the AD mice. Moreover, THz waves treatment effectively attenuated mitochondrial impairment, neuroinflammation, and neuronal loss in the AD mouse brain. Our findings reveal previously unappreciated beneficial effects of THz waves treatment in AD and suggest that THz waves may have the potential to be used as a novel therapeutic intervention for this devastating disease.

Serum-based Raman spectroscopic diagnosis of blast-induced brain injury in a rat model

The diagnosis of blast-induced traumatic brain injury (bTBI) is of paramount importance for early care and clinical therapy. Therefore, the rapid diagnosis of bTBI is vital to the treatment and prognosis in clinic. In this paper, we reported a new strategy for label-free bTBI diagnosis through serum-based Raman spectroscopy. The Raman spectral characteristics of serum in rat were investigated at 3 h, 24 h, 48 h and 72 h after mild and moderate bTBIs. It has been demonstrated that both the position and intensity of Raman characteristic peaks exhibited apparent differences in the range of 800-3000cm^-1 compared with control group. It could be inferred that the content, structure and interaction of biomolecules in the serum were changed after blast exposure, which might help to understand the neurological syndromes caused by bTBI. Furthermore, the control group, mild and moderate bTBIs at different times (a total of 9 groups) were automatically classified by combining principal component analysis and four machine learning algorithms (quadratic discriminant analysis, support vector machine, k-nearest neighbor, neural network). The highest classification accuracy, sensitivity and precision were up to 95.4%, 95.9% and 95.7%. It is suggested that this method has great potential for high-sensitive, rapid, and label-free diagnosis of bTBI.

Terahertz time-domain attenuated total reflection spectroscopy integrated with a microfluidic chip

The integration of a microfluidic chip into terahertz time-domain attenuated total reflection (THz TD-ATR) spectroscopy is highly demanded for the accurate measurement of aqueous samples. Hitherto, however little work has been reported on this regard. Here, we demonstrate a strategy of fabricating a polydimethylsiloxane microfluidic chip (M-chip) suitable for the measurement of aqueous samples, and investigate the effects of its configuration, particularly the cavity depth of the M-chip on THz spectra. By measuring pure water, we find that the Fresnel formulae of two-interface model should be applied to analyze the THz spectral data when the depth is smaller than 210 μm, but the Fresnel formula of one-interface model can be applied when the depth is no less than 210 μm. We further validate this by measuring physiological solution and protein solution. This work can help promote the application of THz TD-ATR spectroscopy in the study of aqueous biological samples.

Detection of biomarkers using terahertz metasurface sensors and machine learning

To achieve classification and concentration detection of cancer biomarkers, we propose a method that combines terahertz (THz) spectroscopy, metasurface sensors, and machine learning. A metasurface sensor suitable for biomarker detection was designed and fabricated with five resonance frequencies in the range of 0.3-0.9 THz. We collected biomarkers of five types and nine concentrations at 100 sets of time-domain spectra per concentration. The spectrum is processed by noise reduction and fast Fourier transform to obtain the frequency-domain spectrum. Five machine learning algorithms are used to analyze time- and frequency-domain spectra and ascertain which algorithm is more suitable for the classification of the biomarker THz spectrum. Experimental results show that random forest can better distinguish five biomarkers with an accuracy of 0.984 for the time-domain spectrum. For the frequency-domain spectrum, the support vector machine performs better, with an accuracy of 0.989. For biomarkers at different concentrations, we used linear regression to fit the relationship between biomarker concentration and frequency shift. Experimental results show that machine learning can distinguish different biomarker species and their concentrations by the THz spectrum. This work provides an idea and data processing method for the application of THz technology in biomedical detection.

Discovering Glioma Tissue through Its Biomarkers' Detection in Blood by Raman Spectroscopy and Machine Learning

The most commonly occurring malignant brain tumors are gliomas, and among them is glioblastoma multiforme. The main idea of the paper is to estimate dependency between glioma tissue and blood serum biomarkers using Raman spectroscopy. We used the most common model of human glioma when continuous cell lines, such as U87, derived from primary human tumor cells, are transplanted intracranially into the mouse brain. We studied the separability of the experimental and control groups by machine learning methods and discovered the most informative Raman spectral bands. During the glioblastoma development, an increase in the contribution of lactate, tryptophan, fatty acids, and lipids in dried blood serum Raman spectra were observed. This overlaps with analogous results of glioma tissues from direct Raman spectroscopy studies. A non-linear relationship between specific Raman spectral lines and tumor size was discovered. Therefore, the analysis of blood serum can track the change in the state of brain tissues during the glioma development.

Review: Emerging Eye-Based Diagnostic Technologies for Traumatic Brain Injury

The study of ocular manifestations of neurodegenerative disorders, Oculomics, is a growing field of investigation for early diagnostics, enabling structural and chemical biomarkers to be monitored overtime to predict prognosis. Traumatic brain injury (TBI) triggers a cascade of events harmful to the brain, which can lead to neurodegeneration. TBI, termed the "silent epidemic" is becoming a leading cause of death and disability worldwide. There is currently no effective diagnostic tool for TBI, and yet, early-intervention is known to considerably shorten hospital stays, improve outcomes, fasten neurological recovery and lower mortality rates, highlighting the unmet need for techniques capable of rapid and accurate point-of-care diagnostics, implemented in the earliest stages. This review focuses on the latest advances in the main neuropathophysiological responses and the achievements and shortfalls of TBI diagnostic methods. Validated and emerging TBI-indicative biomarkers are outlined and linked to ocular neuro-disorders. Methods detecting structural and chemical ocular responses to TBI are categorised along with prospective chemical and physical sensing techniques. Particular attention is drawn to the potential of Raman spectroscopy as a non-invasive sensing of neurological molecular signatures in the ocular projections of the brain, laying the platform for the first tangible path towards alternative point-of-care diagnostic technologies for TBI.

Measurement of tissue optical properties in a wide spectral range: a review [Invited]

A distinctive feature of this review is a critical analysis of methods and results of measurements of the optical properties of tissues in a wide spectral range from deep UV to terahertz waves. Much attention is paid to measurements of the refractive index of biological tissues and liquids, the knowledge of which is necessary for the effective application of many methods of optical imaging and diagnostics. The optical parameters of healthy and pathological tissues are presented, and the reasons for their differences are discussed, which is important for the discrimination of pathologies and the demarcation of their boundaries. When considering the interaction of terahertz radiation with tissues, the concept of an effective medium is discussed, and relaxation models of the effective optical properties of tissues are presented. Attention is drawn to the manifestation of the scattering properties of tissues in the THz range and the problems of measuring the optical properties of tissues in this range are discussed. In conclusion, a method for the dynamic analysis of the optical properties of tissues under optical clearing using an application of immersion agents is presented. The main mechanisms and technologies of optical clearing, as well as examples of the successful application for differentiation of healthy and pathological tissues, are analyzed.

Machine Learning Enhanced Optical Spectroscopy for Disease Detection

Optical spectroscopy plays an important role in disease detection. Improving the sensitivity and specificity of spectral detection has great importance in the development of accurate diagnosis. The development of artificial intelligence technology provides a great opportunity to improve the detection accuracy through machine learning methods. In this Perspective, we focus on the combination of machine learning methods with the optical spectroscopy methods widely used for disease detection, including absorbance, fluorescence, scattering, FTIR, terahertz, etc. By comparing the spectral analysis with different machine learning methods, we illustrate that the support vector machine and convolutional neural network are most effective, which have potential to further improve the classification accuracy to distinguish disease subtypes if these machine learning methods are used. This Perspective broadens the scope of optical spectroscopy enhanced by machine learning and will be useful for the development of disease detection.

Low-Frequency Vibrational Spectroscopy Characteristic of Pharmaceutical Carbamazepine Co-Crystals with Nicotinamide and Saccharin

The pharmaceutical co-crystal has attracted increasing interest due to the improvement of physicochemical properties of active pharmaceutical ingredients. The characterization of pharmaceutical co-crystal is an integral part of the pharmaceutical field. In this paper, the low-frequency vibrational properties for carbamazepine co-crystals with nicotinamide and saccharin (CBZ-NIC and CBZ-SAC) have been characterized by combining the THz spectroscopy with low-wavenumber Raman spectroscopy. The experiment results show that, compared with the individual constituents, CBZ-NIC and CBZ-SAC co-crystals not only have different characteristic absorption peaks in the 0.3-2.5 THz region, but also have significant low-wavenumber Raman characteristic peaks in 0–100 cm⁻¹. Density functional theory was performed to simulate the terahertz and low-wavenumber Raman spectra of the two co-crystals, where the calculation agreed well with the measured vibrational peak positions. The vibrational modes of CBZ-NIC and CBZ-SAC co-crystals were assigned through comparing theoretical results with the experimental spectra. Meanwhile, the low-frequency infrared and/or Raman active of characteristic peaks for such co-crystals were discussed. The results indicate the combination of THz spectroscopy and low-wavenumber Raman spectroscopy can provide more comprehensive low-frequency vibrational information for pharmaceutical co-crystals, such as collective vibration and skeleton vibration, which could play an important role in pharmaceutical science.

Rapid Identification of Easily-Confused Mineral Traditional Chinese Medicine (TCM) Based on Low-Wavenumber Raman and Terahertz Spectroscopy

With the unique advantages of mineral TCMs gradually emerging in clinical treatment, health care, and precaution, it has played an important role in the international medical market. Commonly, mineral TCMs with similar appearance and different processing methods have different effects, but they are easy to be confused in preparation, storage, transportation, and other links, which affects the use and causes related problems. In this paper, six kinds of easily confused mineral TCMs, including coral skeleton, ophicalcitum, calamine, matrii sulfas exsiccatus, gypsum, and alumen, are rapidly characterized using Raman spectroscopy, which can be distinguished with different Raman peaks at 0–300 cm−1 due to the different lattice structure. The THz spectra of these mineral TCMs show that different mineral TCMs have different THz absorption coefficients at 0.3–2.0 THz. Furthermore, compared with the ineffectiveness of the Raman spectrum for differentiating mineral TCMs prepared with disparate processing methods, the terahertz absorption spectrum plays an active role in making up the limitation of low-wavenumber Raman spectroscopy. The combination of low-wavenumber Raman and THz spectroscopy provides a simple and feasible scheme for the identification of mineral TCMs, which could play an important role in the quality control of mineral TCMs.

Terahertz Spectroscopic Analysis in Protein Dynamics: Current Status

Proteins play a key role in living organisms. The study of proteins and their dynamics provides information about their functionality, catalysis and potential alterations towards pathological diseases. Several techniques are used for studying protein dynamics, e.g., magnetic resonance, fluorescence imaging techniques, mid-infrared spectroscopy and biochemical assays. Spectroscopic analysis, based on the use of terahertz (THz) radiation with frequencies between 0.1 and 15 THz (3–500 cm−1), was underestimated by the biochemical community. In recent years, however, the potential of THz spectroscopy in the analysis of both simple structures, such as polypeptide molecules, and complex structures, such as protein complexes, has been demonstrated. The THz absorption spectrum provides some information on proteins: for small molecules the THz spectrum is dominated by individual modes related to the presence of hydrogen bonds. For peptides, the spectral information concerns their secondary structure, while for complex proteins such as globular proteins and viral glycoproteins, spectra also provide information on collective modes. In this short review, we discuss the results obtained by THz spectroscopy in the protein dynamics investigations. In particular, we will illustrate advantages and applications of THz spectroscopy, pointing out the complementary information it may provide.

Temperature dependent terahertz spectroscopy and imaging of orthotopic brain gliomas in mouse models

Terahertz (THz) spectroscopy and imaging were used to differentiate brain gliomas in a mouse model at different temperatures. The THz spectral difference between brain glioma and normal brain tissues at -10°C and 20°C was obtained in the 0.4-2.53 THz range. The absorption coefficient and refractive index values varied with both temperature and frequency. The fresh ex vivo brain glioma tissues were mapped by THz attenuated total reflection (ATR) imaging at 2.52 THz in the temperature range from -20°C to 35°C. Compared with histological examination, THz-ATR imaging could better display the tumor areas at a higher temperature. And the averaged reflectivity of normal tissue was increased with the increase of temperature, whereas the tumor region showed a decreasing trend. Thus, the larger THz imaging difference between glioma and normal tissues could be obtained. Moreover, in vivo brain gliomas in mouse models could also be differentiated clearly from normal brain tissues using THz-ATR imaging at 2.52 THz under room temperature. The THz-ATR images corresponded well with those of visual and hematoxylin and eosin (H&E) stained images. Therefore, this pilot study demonstrated that temperature dependence THz spectroscopy and imaging are helpful to the brain gliomas in mouse model detection.

Advances of terahertz technology in neuroscience: Current status and a future perspective

Terahertz (THz) waves are ranged between microwave and infrared region in the electromagnetic spectrum. THz technology has been demonstrated promising potential for biomedical applications. Exploration of biological effects of THz waves has emerged as a critical new area in life sciences. It is critical to uncover the effects of THz waves on complex biological systems in order to lay out the framework for THz technology development and future applications. Specifically, THz radiation has been shown to affect the nervous system, including the structure of nerve cell membranes, genes expressions, and cytokines level. In this review, we primarily discuss the biological impacts and mechanisms of THz waves on the nervous system at the organisms, cellular, and molecular levels. The future application perspectives of THz technologies in neuroscience are also highlighted and proposed.

Cellular effects of terahertz waves

SIGNIFICANCE

An increasing interest in the area of biological effects at exposure of tissues and cells to the terahertz (THz) radiation is driven by a rapid progress in THz biophotonics, observed during the past decades. Despite the attractiveness of THz technology for medical diagnosis and therapy, there is still quite limited knowledge about safe limits of THz exposure. Different modes of THz exposure of tissues and cells, including continuous-wave versus pulsed radiation, various powers, and number and duration of exposure cycles, ought to be systematically studied.

AIM

We provide an overview of recent research results in the area of biological effects at exposure of tissues and cells to THz waves.

APPROACH

We start with a brief overview of general features of the THz-wave-tissue interactions, as well as modern THz emitters, with an emphasis on those that are reliable for studying the biological effects of THz waves. Then, we consider three levels of biological system organization, at which the exposure effects are considered: (i) solutions of biological molecules; (ii) cultures of cells, individual cells, and cell structures; and (iii) entire organs or organisms; special attention is devoted to the cellular level. We distinguish thermal and nonthermal mechanisms of THz-wave-cell interactions and discuss a problem of adequate estimation of the THz biological effects' specificity. The problem of experimental data reproducibility, caused by rareness of the THz experimental setups and an absence of unitary protocols, is also considered.

RESULTS

The summarized data demonstrate the current stage of the research activity and knowledge about the THz exposure on living objects.

CONCLUSIONS

This review helps the biomedical optics community to summarize up-to-date knowledge in the area of cell exposure to THz radiation, and paves the ways for the development of THz safety standards and THz therapeutic applications.

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.