Open Access
Add to BibSonomy
Abstract
Dental caries is a widespread chronic infectious disease which may induce a series of oral and general problems if untreated. As a result, early diagnosis and follow-up following radiation-free dental caries therapy are critical. Terahertz (THz) waves with highly penetrating and non-ionizing properties are ideally suited for dental caries diagnosis, however related research in this area is still in its infancy. Here, we successfully observe the existence of THz birefringence phenomenon in enamel and demonstrate the feasibility of utilizing THz spectroscopy and birefringence to realize caries diagnosis. By comparing THz responses between healthy teeth and caries, the transmitted THz signals in caries are evidently reduced. Concomitantly, the THz birefringence is also unambiguously inhibited when caries occurs due to the destruction of the internal hydroxyapatite crystal structure. This THz anisotropic activity is position-dependent, which can be qualitatively understood by optical microscopic imaging of dental structures. To increase the accuracy of THz technology in detecting dental caries and stimulate the development of THz caries instruments, the presence of significant THz birefringence effect induced anisotropy in enamel, in combination with the strong THz attenuation at the caries, may be used as a new tool for caries diagnosis.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Dental caries mostly starts inconspicuously in enamel. The mineral part of teeth is calcium-deficient carbonated hydroxyapatite [1], which accounts for 85% of enamel and is crystallized. The basic structure of enamel is the enamel rod, an elongated columnar structure made up of millions of hydroxyapatite crystals. The inorganic material present in dentin is also in the form of hydroxyapatite crystals, but the crystal size is much smaller than that in enamel. Caries is a chronic infectious disease that results in demineralization of the tooth’s structures, decomposition of organic matter and consequently chronic progressive deterioration, due to a combination of factors including bacterial infection. Enamel caries refers to carious lesions that develop in the enamel, which is difficult for dentists to diagnosis. Untreated enamel caries will progress to dentin caries, tooth loss, biting problems, and potentially nutrition and metabolism disorders [2]. Therefore early diagnosis of enamel caries is critical [3]. Radiography, which is frequently used in clinical practice to assess dental cavities, generates produces harmful ionization. Although the low radiation dosage for healthy, it can still cause concern in some patients, particularly youngsters and pregnant women. In addition, several new optical-based methods for detecting dental caries have been proposed [4], including quantitative light fluorescence (QLF), multi-photon imaging, infrared thermography, optical coherence tomography (OCT), polarized Raman spectroscopy [5], etc. Although various advanced technologies for caries diagnosis are being developed, non-invasive, radiation-free and high contrast detection methods are still in urgent need.
THz waves correspond to the part of the electromagnetic spectrum with frequencies ranging from 0.1 THz to 10 THz, and the wavelengths ranging from 3 mm to 30 µm. Its low photon energy, non-ionizing radiation, water sensitivity and high penetrability makes it very useful for biomedical sensing. To this goal, two major approaches including spectroscopy and imaging have been established. Using spectroscopy, THz absorption properties of amino acid, polypeptide [6], DNA [7] and proteins [8] have been characterized, cancer [9] and tumor tissues [10] can be screened, blood glucose [11] can also be noninvasive monitored. THz imaging aids in the diagnosis of many types of cancer or tumor tissues such as liver cancer [12], breast cancer [13], skin cancer [14], brain tumor [15], etc. It can also be used to detect skin burns [16] and the effects of skin moisturization [17], to monitor the healing of skin scars [18] and to detect bone tissues which were damaged by exposure to hydrofluoric acid [19]. Furthermore, it has been shown that strong THz pulses cause rapid dissociation of short dsDNA, paving the path for the design of a new range of functional DNA nanomaterials [20].
Dentistry is also a critical component of THz biomedicine. Carious tissue typically has a greater THz transmission attenuation than healthy enamel and that the carious tissue may not absorb THz waves rather than scattering [21]. In addition, the THz refractive index of enamel is higher than that of dentin [22]. Based on these variations, it has been claimed that distinguishing carious enamel and healthy enamel by transmission or reflection imaging is possible [21]. THz time-of-flight imaging has been verified to distinguish enamel and dentin. Three-dimensional structure of tooth were imaged, and the enamel thickness was measured accurately and reliably by THz pulses [23]. THz pulse imaging also exhibited capability for determining the depth of artificial acid gel demineralization across a limit range. As a result, THz pulses can be promisingly used to measure the depth of lesion [24]. In addition to using the time-domain information of THz pulse, the frequency-domain information can also be used for three-dimensional imaging. The results of THz pulse imaging are similar to those of X-ray imaging, demonstrating the effectiveness of THz pulse imaging for dental caries measurement [25]. A miniaturized optical fiber-coupled THz endoscope system for oral inspection has also been developed [26]. Although THz technology has been studied in dentistry, compared with other medical applications, more comprehensive and systematic research is required on the mechanism of THz wave-tooth interaction.
In this work, we have measured the transmitted THz signals of carious enamel, healthy enamel and dentin in tooth sections using THz time-domain spectroscopy (THz-TDS) and implemented azimuthal angle dependent experiments. To the best of our knowledge, we have observed for the first time that enamel is birefringent in the THz frequency region and correlates with the direction of growth of hydroxyapatite crystals within the enamel. There is significant attenuation in carious enamel compared to healthy enamel, and the THz birefringence is less pronounced, which may be related to the degradation of the internal crystal structure of the carious enamel. No significant birefringence is detected at the dentin. We think that the significance of the THz birefringence phenomenon could provide a new early diagnosis technology for THz caries testing, hence advancing the use of THz technology in biomedicine.
2. Experimental setup and sample preparation
2.1 THz-TDS
The experiments are carried out on a home-built THz-TDS illustrated in Fig. 1(a). The system is driven by a Ti:Sapphire femtosecond laser oscillator with a central wavelength of 800 nm, a pulse duration of 70 fs, a repetition rate of 80 MHz and a maximum output power of 1.1 W. The laser pulses are divided into a pump light and a probe beam by a beam splitter, and the pump light passes through a translation stage and is focused onto a low-temperature grown GaAs photoconductive antenna for THz generation. The radiated THz pulses are collimated and then focused onto the tooth sample by two off-axis parabolic mirrors, while the transmitted THz signal is collimated and focused onto a ZnTe detecting crystal in conjunction with the probe light. Electro-optic sampling including a quarter-wave plate, a Wollaston prism and two photo diodes is employed to coherently record THz temporal waveforms, and the sample response in THz frequency range can be extracted. The part of the optical path containing THz light is sealed and purged with nitrogen gas to eliminate the effect of water vapor in the air. The relative humidity of the test environment is maintained at 1.5%, and the temperature is 23°. In our experiment, we clamped the tooth sections using a metal aperture with a diameter of ∼1 mm to limit the size of the THz beam spot that interacts with the samples.
Fig. 1. Schematic diagram of the experimental setup and tooth sections. (a) THz-TDS. (b) and (c) The vertical section photos for a healthy tooth and a caries, respectively.
2.2 Tooth sample preparation
Two tooth samples (one with caries and one without) measured in THz-TDS are sliced to vertical sections. Prior to that, the teeth are first soaked in a solution of 3% hydrogen peroxide, and then are embedded by methyl methacrylate. Following the procedural of ultraviolet curing procedure, the teeth are sliced to vertical sections with various thickness from ∼200 µm-∼2 mm. Typical optical photos acquired with a cellphone (HUAWEI p30) for the healthy tooth sections and carious sample are shown in Fig. 1(b) and Fig. 1(c), respectively. It can be seen from these photos that the healthy tooth section contains healthy enamel and healthy dentin, while the carious section manifests carious enamel which is denoted by red dashed rectangle.
Tooth sections are sterilized and prevented from growing bacteria and fungus by soaking them in a 3% hydrogen peroxide solution. Prior to conducting THz spectroscopy testing, the tooth sections are taken out from the 3% hydrogen peroxide solution, sucked the solution on the surface of the sections with absorbent paper, and then left to dry in air for 20-30 minutes to ensure the surfaces of the tooth sections are dry. Following these pretreatments, the sections are placed in the THz-TDS sample holder.
3. Results
3.1 Distinguish caries from healthy teeth via THz spectroscopy
Figure 2(a) depicts the THz time-domain waveforms for healthy enamel, dentin, carious enamel and the reference signal, respectively. These transmitted THz signals are gathered at several positions on the same caries section to ensure that the section thickness remains constant. The caries section is ∼350 µm thick. The transmitted signals for healthy enamel, dentin and carious enamel show significant attenuation, and the transmitted THz signal for the carious enamel is ∼40% of that for the healthy enamel. The transmitted signal for dentin is the strongest among the three tissues. The THz phase delay for the enamel is 2.65 ps while that for dentin is 1.7 ps when both of them compare with the peak signal of the reference. Using fast Fourier transform (FFT) to calculate the transmitted spectra, as depicted in Fig. 2(b), the half bandwidth of the measured tooth tissue transmitted spectra are narrowed from 1.65 THz for the reference signal to 0.81 THz for the healthy enamel, 0.79 THz for the dentin, and 0.72 THz for the carious enamel, respectively. Their high cut-off frequencies are ∼1.85 THz for the dentin, ∼1.71 THz for the healthy enamel ∼1.61 THz for the carious enamel, implying the possibility for distinguish caries from healthy teeth via THz spectroscopy.
Fig. 2. THz spectra for healthy enamel, dentin and carious enamel. (a) The reference signal and the transmitted THz electric fields signal for enamel, caries and dentine, and (b) their corresponding Fourier transform spectra. Their transmissivity spectra, relative loss spectra and refractive index spectra are illustrated in (c), (d) and (e).
Besides amplitude spectrum differences, we further calculate their transmissivity and relative loss spectra with the following Eqs. (1) and (2).
where ${A_t}(\omega )$ and ${A_{ref}}(\omega )$ denote the transmitted amplitude spectra for the sample and the reference, respectively, $\omega $ is the angular frequency. The calculated transmissivity spectrum is illustrated in Fig. 2(c). In the frequency range of 0.2 THz-2.0 THz, has the lowest transmissivity, while the healthy enamel transmissivity is higher than that of the carious enamel. The average transmissivity of dentin is slightly higher than that of healthy enamel. With increasing frequency, the transmissivity of the three dental tissues reduce rapidly. The relative loss reflects the ability of dental tissue to absorbed, reflected and scattered THz waves, as illustrated in Fig. 2(d). Carious enamel exhibits strong loss for THz waves when comparing with the healthy enamel and dentin, and the relative loss of the three dental tissues have the tendency of stronger loss along with the increasing of the frequency.We also calculate the refractive indices for both the enamel and the dentin $n(\omega )$ with following equation:
where $\varphi (\omega )$ is the phase difference between the sample signal and the reference signal, c denotes the light speed in vacuum, d represents the sample thickness. Their refractive indices spectra are illustrated in Fig. 2(e). Between 0.2 THz-2.0 THz, the average refractive index is ∼2.9-3.0 for the enamel and ∼2.4 for the dentin.3.2 THz anisotropy of healthy enamel
Along with the difference in THz transmission spectra between healthy enamel and caries, more interestingly, we also find that healthy enamel has THz anisotropy. The healthy tooth section thickness is ∼430 µm. To eliminate the effect of birefringence of electro-optic (ZnTe) crystal, we replace the THz source and detector with a pair of InGaAs photoconductive antennas driven by a femtosecond fiber laser. The linearly polarized THz pulses in the horizontal direction is normally incident onto the sample surface, and the detector only records horizontally polarized THz pulses. From now on, all these measurements are in ambient air. The transmitted THz signal for the enamel is azimuthal angle dependent. The specified detection position is indicated by the black dashed circle in Fig. 1(b). The transmitted signals are summarized in Fig. 3, when the tooth section is rotated clockwise from 0° to 165° (with a step of 15°), respectively. When the rotation angle is 0°, the tooth section is positioned in the sample holder in the orientation illustrated in Fig. 1(b). Enamel demonstrates obvious anisotropy. As the azimuthal angle changes, the delay time of the THz temporal waveform peak value varies explicitly. The two different peaks can be seen in some of the subplots, and the variation in peak-to-peak values is also significant. The red dashed line shows the peak position for the first peak, while the blue for the second peak of the transmitted signals, from which the time difference between the two peaks can be easily discerned. This phenomenon implies the presence of THz birefringence in enamel.
Fig. 3. Transmitted THz signals for the healthy enamel, corresponding to the position in Fig. 1(b). The red dashed lines are for the first peak value while the blue for the second. (a)-(l) The variation process for the transmitted THz temporal waveforms with respect to different azimuthal angles from 0° to 165° with 15° internal.
The tooth is a kind of spatially nearly axially symmetrical organ, and its THz anisotropy may be spatially distributed. The THz anisotropy considering both temporal waveform peak-to-peak values and the equivalent refractive index for eight different locations uniformly selected in the tooth enamel are plotted in Fig. 4. The equivalent refractive index n is calculated with the following equation:
where $\Delta t$ is the THz temporal waveform peak value time difference between the sample signal and the reference.Fig. 4. THz anisotropy dependent on the location of healthy enamel. (a) Vertical section photo of the healthy tooth, (b)-(i) THz peak-to-peak values and the equivalent refractive indices at different locations in the enamel, taken from the black dashed circles in a), respectively.
The azimuthal angle anisotropy of the peak-to-peak value and equivalent refractive index of the transmitted signal at the enamel is illustrated in Fig. 4. In the center of Fig. 4 (Fig. 4(a)), we present the photo for the healthy tooth vertical section which is the same as shown in Fig. 1(b). This is the orientation of the section for the data at 0° in Fig. 4(b)-(i). The tooth section was rotated 360° clockwise and one set of data was collected at every 15° internal rotation. Therefore, 24 sets of data are collected at each test location and plotted as an azimuthal angle dependence graph, where the peak-to-peak value is plotted by the red dashed line and the equivalent refractive index is plotted by the blue solid line. The eight selected positions are marked with black dashed circles in Fig. 4(a).
As the tooth section is rotated, there is a clear change in the THz peak-to-peak values of the enamel. For example, in Fig. 4(b), there is a long axial direction (rotated by 315°/330° and 135°/150°), defined as “long axis direction”, which represents that when the tooth section rotates by this direction, the THz peak-to-peak value can achieve a local maximum. The peak-to-peak value drops as the tooth section rotates, reaching to a local minimum, and then increasing until it reaches the local maximum again. This second local maximum represents a secondary long axis (rotated by 45°/60° and 225°/240°), which is referred to as the “secondary long axis direction”. The angle between the long axis direction and the secondary long axis is ∼90°, which leads to the pattern like a “four-petalled flower in full bloom”. The global peak value reaches the maximum at 315°, while the minimum appears at 15°. The maximum value is about three times higher than the minimum. The equivalent refractive indices also show obvious anisotropy and have amazing correlation with the anisotropy of peak-to-peak value. The equivalent refractive indices have two values of ∼2.6 and ∼3.0. The azimuthal angle corresponding to the larger refractive indices is always consistent with the “short petal” in the peak-to-peak anisotropy diagram.
When the detection position varies, as illustrated in Fig. 4(b)-(e), the THz anisotropy in the left half-side of the enamel changes. The anisotropy diagrams show a counterclockwise rotation. Similarly, for the right half-side in Fig. 4(f)-(i), the variation tendency also has a counterclockwise rotation behavior. This position dependence further suggests that the THz anisotropy is highly connected with the internal tissue structure of the enamel, which is a biological tissue with a regular position-dependent internal texture. However, no significant anisotropy is found in the transmitted signal at the dentin (see Supplemental document). In practical applications, a proper choice of detection azimuth may be beneficial for the enamel caries diagnosis at the enamel. The THz anisotropy phenomenon can be employed to keep the detection azimuth in the direction of the maximum THz transmissivity, hence lowering the detection instrument's output power consumption.
3.3 Microscopic origin of the THz anisotropy for healthy enamel
Since the THz anisotropic regularity depends the enamel position, we hypothesize that it is related to the internal growth structure of the tooth. To verify this suspicion, optical microscopic imaging of the enamel is measured. Figure 5(a) illustrates the optical microscopic image under a 50 times magnification. It shows the presence of three kind of significant textural structures in the enamel, which are indicated by yellow arrows. The elongated radial structure indicated by arrow 1 is the enamel rod, which is the basic structure to make up the enamel. The structure indicated by arrow 2 is the enamel incremental line, also known as the line of Retzius. Enamel incremental lines are generated in the direction of enamel formation and are developmental interstitial lines. Along with the two aforementioned textural structures present within the enamel itself, there is also a texture indicated by arrow 3, which does not change direction with position, manifesting the knife cutting edge left during section processing.
Fig. 5. The correlation between optical microscopic images and THz anisotropy for healthy enamel. (a) Three typical textural structures at enamel: yellow line 1 for enamel rods; line 2 for enamel incremental lines; line 3 for knife cutting edge, respectively. (b) and (c) Microscopic image and THz anisotropy corresponding to Fig. 4(c). (d) and (e) Microscopic image and THz anisotropy corresponding to Fig. 4(h).
The following part will discuss the comprehensive comparation between THz anisotropy for the enamel and its microstructures. Figure 5(b) and Fig. 5(c) give the microscopic image and its corresponding anisotropy. It can be seen that when we rotate the tooth section, the enamel rod direction perpendicular to the incident THz polarization is the secondary long axis direction. While for the parallel case, it is the long axis direction. The peak-to-peak value reaches local minimum when the enamel rod direction is at an azimuthal angle of 45° with respect to the THz polarization. The fixed directional texture caused by the knife cutting edge does not exhibit an anisotropy behavior. Such phenomenon is also observed in another enamel position as shown in Fig. 5(d) and Fig. 5(e). All these findings suggest periodic texture structure of the enamel rods is the microscopic mechanism underlying THz anisotropy.
3.4 Caries diagnosis via monitoring THz anisotropy variation
The textural structures in the enamel are responsible for the THz anisotropy. In the case of enamel caries, the internal crystalline tissue at the site of the caries is destroyed. Figure 6(a) illustrates a vertical section containing enamel caries. The two locations in enamel caries are circled on the right of the three black dashed circles. Anisotropy of the healthy enamel is drawn in Fig. 6(b), while those for the carious enamel are ploted in Fig. 6(c) and Fig. 6(d). The refractive index difference between the long axis direction and the secondary long axis direction at the carious enamel decreases from 0.25 in Fig. 6(b) (healthy enamel) to 0.19 in Fig. 6(c) and 0.14 in Fig. 6(d). The THz attenuation also increases, as does the pattern's weakness. Therefore, by carefully monitoring the variation of THz anisotropic signal, we can infer the presence and severity of enamel caries.
Fig. 6. THz anisotropy dependent on the location of the enamel caries. (a) Vertical section photo for the caries. THz anisotropy of the peak-to-peak value and the equivalent refractive index for (b) the healthy enamel, and (c), (d) for the carious enamel.
4. Discussion
The mechanism for the azimuthal anisotropy caused by THz birefringence in enamel can be explained as following. The refractive index angle dependence anisotropy show that the refractive index of the enamel varies in two mutually orthogonal directions, one of which is the fast light direction while the other is the slow light direction. The optical axis direction is the growth direction of the enamel rods, roughly parallel to the surface of the dental section. Figure 7 illustrates THz birefringence in enamel. When the enamel rods orientation is perpendicular to the THz polarization, e.g. the tooth section is rotated by 30° in Fig. 5(c), corresponding to Fig. 7(a), the slow light direction is parallel to the THz polarization. The linearly polarized THz pulses are fully slanted into the slow light direction, resulting in significant observed signals. Similarly, when the enamel rods orientation is parallel to the THz polarization, e.g. the tooth section is rotated by 120° in Fig. 5(c), corresponding to Fig. 7(b), the fast light direction is parallel to the THz polarization, the linearly polarized THz pulses are fully converted into the fast light direction. However, when neither the fast nor the slow light directions of the enamel rods are parallel to the direction of THz polarization, typically the tooth section is rotated by 75° as in Fig. 5(c), corresponding to Fig. 7(c), the angle between the fast or slow light directions and the THz polarization are both 45°. The light intensity of the incident THz pulses is separated into the fast and slow directions. At this time, due to the different refractive indices of the two directions, the propagation speed of THz pulses is different. The phase difference between the two directions of THz pulses appears, and the linearly polarized THz pulses are transformed into elliptical polarized.
Fig. 7. THz birefringence in enamel. The enamel rods orientation is perpendicular (a) and parallel (b) to the THz polarization. The linearly polarized THz pulses are fully tilted into the slow light direction (a) or fast light direction (b), respectively. (c) Neither the fast nor the slow light directions of the enamel rods are parallel to the direction of THz polarization, the light intensity of the incident THz pulses is divided into the fast and slow directions. The detector can only detect THz pulses with horizontal polarization.
Using the foregoing theory, we can further comprehend the experimental phenomenon in Fig. 3 in the time domain. The red dashed line in Fig. 3 can be further read as the delay time of the THz signal in the fast light direction of the enamel, and the blue dashed line denotes the slow light direction. In Fig. 3(a)-(c), the slow light direction of the enamel is roughly parallel to the THz polarization, the delay time of THz temporal waveform peak value is late. On the contrary, in Fig. 3(g)-(i), the fast light direction of the enamel is roughly parallel to the THz polarization, the delay time is early. In the foregoing instances, the incident THz pulses are concentrated in one axial direction and therefore the signal peak value is large. Figure 3(d)-(f) and Fig. 3(j)-(l) show the transition azimuthal angles which neither the fast nor the slow light direction of enamel is parallel to the THz polarization, the incident THz pulses are split into two directions and the transmitted signal peak value is low. In particular, as can be seen in Fig. 3(h), the delay time of THz temporal waveform valley value of the fast light is close to the peak value of the slow light, so that in Fig. 3(f) and Fig. 3(j) the two transmitted signals in the enamel are superimposed together, where the peak of the slow light does not appear.
Using thicker sections is beneficial to separate the fast and slow light. We test a ∼670-µm thick healthy tooth section, and repeat the azimuthal angle dependent experiment. The transmitted signals at each azimuthal angle are illustrated in Fig. 8. Figure 8(b) and 8(h) approximately correspond to the situation described in Fig. 7(a) and 7(b), respectively, while Fig. 8(e) and 8(h) correspond to the situation described in Fig. 7(c). Due to the thicker section, the delay time of THz temporal waveform valley value of the fast light do not coincide with the peak value of the slow light, as illustrated in Fig. 8(k). This observation indicates enhanced thickness can improve the THz birefringence. A THz polarizer is inserted between the sample and the detector. When the azimuthal angle is 150°, rotating the polarizer makes the maximum THz electric field at 45° and -45° with respect to the horizon direction. The detected signal is shown in Fig. 8(m) and 8(n). The overlapping fast and slow light in Fig. 8(k) are well separated by a polarizer. The fast light is shown in Fig. 8(m) and the slow light is shown in Fig. 8(n). By adding the two signals together numerically, the result shown in Fig. 8(o) is almost identical to the signal directly measured in Fig. 8(k). This result indicates that THz birefringence does occur in enamel.
Fig. 8. Transmitted THz signals for a 670-µm thick healthy enamel section. The red dashed lines are for the fast light, while the blue for the slow. (a)-(l) The variation process for the transmitted THz temporal waveforms with respect to different azimuthal angles from 0° to 165° with 15° internal. (m) and (n) The detected signals for the maximum THz electric field at 45° and -45° with respect to the horizon direction, when the azimuthal angle is 150°. (o) The combined signal produced by the superposition between (m) and (n) numerically.
The phenomenon of THz birefringence in enamel may be related to the arrangement of the hydroxyapatite crystals within the enamel rods, the structure, size, number of crystals and volume of intercrystalline pores. Under some azimuthal angles, the THz birefringence of enamel may lead to the overlap of fast light and slow light, resulting in drastic changes in the time-domain waveform and spectrum. This may lead to misdiagnosis using THz spectroscopy detection methods. Therefore, we strongly recommend rotating the sample azimuthal angle to avoid the influence of birefringence when using THz technique for caries diagnosis, or using our proposed detection method based on birefringence induced anisotropy.
In our measurements, to avoid the interference from water vapor, we purged the system at first and obtain THz spectroscopic information and then the anisotropic measurements were conducted in ambient air. It implies that there is possibility to apply such technique in significant environment such as the mouth in the future.
5. Conclusion
In this work, we systematically study the THz spectral response of the healthy teeth and caries by THz-TDS. In terms of the attenuation of the THz transmission spectrum, we find that caries has greater THz attenuation than that of the healthy teeth. Furthermore, we also find that there is obvious THz birefringence in the enamel, but this birefringence decreases along with the occurrence and severity development of the dental caries, which is well interpreted through observing their microstructures. We see the anisotropic change from the order of healthy teeth to the disorder of caries by using THz anisotropic spectroscopy technology, which proves that THz technology can be used not only to study life being developing from intangible to tangible [27], but also the order to disorder of the biological tissue. The strong attenuation together with anisotropic disruption of the transmitted THz signal could provide a joint basis for the detection of the location, size and severity of dental caries, improving the accuracy of the detection and may be employed to for developing novel early diagnosis technology for THz caries diagnosis.
Funding
Shenzhen Fundamental Research Program (2021Szvup080).
Acknowledgments
We also thank the National Natural Science Foundation of China (61905007), the National Key R&D Project (2019YFB2203102), and the Open Fund of Guangdong Provincial Key Laboratory of Information Photonics Technology (Guangdong University of Technology, No. GKPT20).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. M. Goldberg, Understanding dental caries (Springer, 2016).
2. Centers for Disease Control and Prevention, “Achievements in public health, 1990-1999: Fluoridation of drinking water to prevent caries,” JAMA-J. Am. Med. Assoc.283(10), 1283–1286 (2000). [CrossRef]
3. K. W. Davidson, M. J. Barry, C. M. Mangione, M. Cabana, A. B. Caughey, E. M. Davis, K. E. Donahue, C. A. Doubeni, M. Kubik, and L. Li, “Screening and interventions to prevent dental caries in children younger than 5 years: US Preventive Services Task Force recommendation statement,” JAMA 326(21), 2172–2178 (2021). [CrossRef]
4. A. Hall and J. Girkin, “A review of potential new diagnostic modalities for caries lesions,” J. Dent. Res. 83(Spe1_suppl C), 89–94 (2004). [CrossRef]
5. A. C.-T. Ko, M. Hewko, M. G. Sowa, C. C. Dong, and B. Cleghorn, “Detection of early dental caries using polarized Raman spectroscopy,” Opt. Express 14(1), 203–215 (2006). [CrossRef]
6. Z.-P. Zheng and W.-H. Fan, “First principles investigation of L-alanine in terahertz region,” J. Biol. Phys. 38(3), 405–413 (2012). [CrossRef]
7. H. Yan, W. H. Fan, Z. P. Zheng, and J. Liu, “Terahertz spectroscopy of DNA nucleobases: cytosine and thymine,” Spectrosc. Spectr. Anal. 33(10), 2612–2616 (2013). [CrossRef]
8. K. M. Tych, A. D. Burnett, C. D. Wood, J. E. Cunningham, A. R. Pearson, A. G. Davies, and E. H. Linfield, “Applying broadband terahertz time-domain spectroscopy to the analysis of crystalline proteins: a dehydration study,” J. Appl. Crystallogr. 44(1), 129–133 (2011). [CrossRef]
9. Y. Q. Cao, P. J. Huang, X. Li, W. T. Ge, D. B. Hou, and G. X. Zhang, “Terahertz spectral unmixing based method for identifying gastric cancer,” Phys. Med. Biol. 63(3), 035016 (2018). [CrossRef]
10. S. Yamaguchi, Y. Fukushi, O. Kubota, T. Itsuji, T. Ouchi, and S. Yamamoto, “Origin and quantification of differences between normal and tumor tissues observed by terahertz spectroscopy,” Phys. Med. Biol. 61(18), 6808–6820 (2016). [CrossRef]
11. O. Cherkasova, M. Nazarov, and A. Shkurinov, “Noninvasive blood glucose monitoring in the terahertz frequency range,” Opt. Quantum Electron. 48(3), 217 (2016). [CrossRef]
12. S. Nakajima, H. Hoshina, M. Yamashita, C. Otani, and N. Miyoshi, “Terahertz imaging diagnostics of cancer tissues with a chemometrics technique,” Appl. Phys. Lett. 90(4), 041102 (2007). [CrossRef]
13. T. Bowman, M. El-Shenawee, and L. K. Campbell, “Terahertz transmission vs reflection imaging and model-based characterization for excised breast carcinomas,” Biomed. Opt. Express 7(9), 3756–3783 (2016). [CrossRef]
14. A. Rahman, A. K. Rahman, and B. Rao, “Early detection of skin cancer via terahertz spectral profiling and 3D imaging,” Biosens. Bioelectron. 82, 64–70 (2016). [CrossRef]
15. S. Yamaguchi, Y. Fukushi, O. Kubota, T. Itsuji, T. Ouchi, and S. Yamamoto, “Brain tumor imaging of rat fresh tissue using terahertz spectroscopy,” Sci. Rep. 6(1), 1–6 (2016). [CrossRef]
16. P. Tewari, N. Bajwa, R. S. Singh, M. O. Culjat, W. S. Grundgest, Z. D. Taylor, C. P. Kealey, D. B. Bennett, K. S. Barnett, and A. Stojadinovic, “In vivo terahertz imaging of rat skin burns,” J. Biomed. Opt. 17(4), 040503 (2012). [CrossRef]
17. L. Martinez-Meza, S. Rojas-Landeros, E. Castro-Camus, and M. Alfaro-Gomez, “Evaluation of skin moisturizer effects using terahertz time domain imaging,” in Photonics in Dermatology and Plastic Surgery 2018(International Society for Optics and Photonics2018), p. 104670S.
18. S. Fan, B. S. Ung, E. P. Parrott, V. P. Wallace, and E. Pickwell-MacPherson, “In vivo terahertz reflection imaging of human scars during and after the healing process,” J. Biophotonics 10(9), 1143–1151 (2017). [CrossRef]
19. W. E. Baughman, H. Yokus, S. Balci, D. S. Wilbert, P. Kung, and S. M. Kim, “Observation of hydrofluoric acid burns on osseous tissues by means of terahertz spectroscopic imaging,” IEEE Trans. Terahertz Sci. Technol. 17(4), 798–805 (2013). [CrossRef]
20. A. A. Greschner, X. Ropagnol, M. Kort, N. Zuberi, J. Perreault, L. Razzari, T. Ozaki, and M. A. Gauthier, “Room-temperature and selective triggering of supramolecular DNA assembly/disassembly by nonionizing radiation,” J. Am. Chem. Soc. 141(8), 3456–3469 (2019). [CrossRef]
21. D. A. Crawley, C. Longbottom, B. E. Cole, C. M. Ciesla, D. Arnone, V. P. Wallace, and M. Pepper, “Terahertz pulse imaging: a pilot study of potential applications in dentistry,” Caries Res. 37(5), 352–359 (2003). [CrossRef]
22. Y. C. Sim, I. Maeng, and J. H. Son, “Frequency-dependent characteristics of terahertz radiation on the enamel and dentin of human tooth,” Curr. Appl. Phys. 9(5), 946–949 (2009). [CrossRef]
23. D. A. Crawley, C. Longbottom, V. P. Wallace, B. E. Cole, D. D. Arnone, and M. Pepper, “Three-dimensional terahertz pulse imaging of dental tissue,” J. Biomed. Opt. 8(2), 303–307 (2003). [CrossRef]
24. E. Pickwell, V. P. Wallace, B. E. Cole, S. Ali, C. Longbottom, R. J. Lynch, and M. Pepper, “A comparison of terahertz pulsed imaging with transmission microradiography for depth measurement of enamel demineralisation in vitro,” Caries Res. 41(1), 49–55 (2007). [CrossRef]
25. K. Kamburoğlu, B. Karagöz, H. Altan, and D. Özen, “An ex vivo comparative study of occlusal and proximal caries using terahertz and X-ray imaging,” Dentomaxillofac. Radiol. 48(2), 20180250 (2019). [CrossRef]
26. Y. B. Ji, E. S. Lee, S.-H. Kim, J.-H. Son, and T.-I. Jeon, “A miniaturized fiber-coupled terahertz endoscope system,” Opt. Express 17(19), 17082–17087 (2009). [CrossRef]
27. H. T. Xiong, J. H. Cai, W. H. Zhang, J. S. Hu, Y. X. Deng, J. G. Miao, Z. Y. Tan, H. Li, J. C. Cao, and X. J. Wu, “Deep learning enhanced terahertz imaging of silkworm eggs development,” iScience 24(11), 103316 (2021). [CrossRef]
