Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jul 1.
Published in final edited form as: Bioelectromagnetics. 2011 Jan 31;32(5):360–366. doi: 10.1002/bem.20650

Dielectric properties of human skin at an acupuncture point in the 50 - 75 GHz frequency range. A pilot study

S J-P Egot-Lemaire 1, MC Ziskin 1
PMCID: PMC3100425  NIHMSID: NIHMS274631  PMID: 21284009

Abstract

The reason for using acupuncture points as exposure sites in some applications of millimeter wave therapy has been unclear. Acupuncture points have been suspected to exhibit particular direct current (DC), low-frequency electrical and optical properties compared to surrounding skin. To assess if such a biophysical correlation could exist at millimeter wave frequencies used in the therapy, we investigated the dielectric properties of an acupuncture point on the forearm skin within the 50 – 75 GHz range. These properties were compared with those of a neighboring ipsilateral control area and a corresponding contralateral control area. The complex reflection coefficient at the end of an open-ended rectangular waveguide loaded with a Teflon plug was measured with a vector network analyzer. A suitable model of the aperture admittance was used to calculate the dielectric properties of the skin at the measured spots. Statistical analyses were conducted with an ANOVA to compare the three sites. From these analyses, the dielectric properties of the acupuncture site were found to be somewhat different from those of surrounding non-acupuncture sites from 50 to about 61 GHz, in the case of the real part of the complex permittivity.

Keywords: complex permittivity, vector network analyzer, open-ended waveguide, reflection coefficient, aperture admittance

INTRODUCTION

Millimeter wave therapy (MMWT) has been developed and extensively utilized in Russia and in several Eastern European countries for more than thirty years [Rojavin and Ziskin, 1998]. MMWT has been credited with being able to treat more than fifty different conditions, including pain, skin wounds, gastrointestinal disorders, as well as pulmonary, cardiovascular, bone and joint diseases [Rojavin and Ziskin, 1998; Pakhomov and Murphy, 2000]. When the disease being treated is surface wounds and skin diseases or arthritis, millimeter waves are applied at the site of the lesion. When the treatment deals with internal organs, some recommended sites of exposure in MMWT are the lower end of the sternum, tender sites on the body surface, or acupuncture points [Chernavsky et al., 1994; Rojavin and Ziskin, 1998]. The reason for using acupuncture points as exposure sites in MMWT has been unclear, as underlined in the review by Usichenko et al. [2006]: “The most intriguing question for future research might be the evaluation of the role of the exposure site, especially in relation to the topography of the acupuncture meridian system”. MMWT has been shown to involve the peripheral nervous system, especially free nerve endings in the skin which appear to be primary targets of millimeter waves [Pakhomov et al., 1997; Radzievsky et al., 2001, 2004]. Also, acupuncture spots have been suggested to have a higher nerve density, potentially within the skin or at muscle interfaces [Chernavsky et al., 1994; Feng and Ifrim, 2004; Li et al., 2004; Zhao 2008], but other reports are conflicting [Wick et al., 2007].

Electrical properties of acupuncture points and meridians at direct current (DC) or low frequencies (up to a few kHz) have been investigated over the past decades and the corresponding studies have recently been critically reviewed by Ahn et al. [2008]. Some of them tended to show an association between acupuncture points and reduced electrical impedance [Reichmanis et al., 1975, 1976; Hyvarinen and Karlsson, 1977; Falk et al., 2000]. However, the review authors considered that to date there has not been sufficient evidence to “conclusively support the claim that acupuncture points or meridians are electrically distinguishable” at these frequencies. Nevertheless, it is stated that “the evidence is more supportive of the claim that acupuncture meridians are electrically distinct”. Besides this probable difference in electrical properties, a difference in optical properties between acupuncture points or meridians compared to surrounding skin has been reported in terms of an increased light absorption and a different light propagation and reflectance [Choi et al., 2003; Yang et al., 2007], either in the infrared or visible domain.

Because the penetration depth of optical frequencies in the body is smaller than at low frequencies, some of these papers suggest that the detected difference between acupuncture points and surrounding sites could exist within the skin. This is precisely where millimeter waves interact with the body. The three common “therapeutic” frequencies employed in MMWT are 42.2 GHz, 53.6 GHz and 61.2 GHz. At these frequencies, the penetration depth of the waves is less than 1 mm [Alekseev et al., 2008], and therefore they are absorbed within the skin. If the skin had particular biophysical features at acupuncture sites, the related hypothetical constituents could possibly interact with millimeter waves, thus being another lead to pursue in an attempt to find a justification for using acupuncture sites in MMWT. Therefore, it is interesting to investigate the biophysical interaction of millimeter waves with the skin at acupuncture sites. As such, dielectric properties are pertinent biophysical parameters to study. The aim of this study was to test whether an acupuncture point could exhibit different dielectric properties than surrounding skin in the frequency range 50 – 75 GHz. This paper addresses this question in a pilot study in humans.

In order to properly characterize the dielectric properties of the skin, the experimental approach of this work was first to measure the millimeter wave complex reflection coefficient at the skin surface with a waveguide-based probe and then implement a suitable probe model to reconstruct the dielectric properties of the skin. This method was used to compare the dielectric properties of an acupuncture site to those of two control sites on several volunteers.

MATERIALS AND METHODS

Measurement of the millimeter wave complex reflection coefficient

The complex reflection coefficient at the skin surface was measured using an open-ended rectangular waveguide technique. We used the state-of-the-art Rohde & Schwarz (Munich, Germany) ZVA50 vector network analyzer equipped with ZVA Z75 millimeter wave extenders to cover the frequency range 50 to 75 GHz. Thus, standard flanged V-band waveguides, also provided by Rohde & Schwarz (size: WR-15; dimensions of the aperture: a = 3.76 mm; b = 1.88 mm), were utilized to perform the measurements. The magnitude and phase of the reflection coefficient were recorded over 201 linearly spaced frequency points; the chosen IF bandwidth was 500 Hz and the output power 0 dBm. The incident power density was estimated to be about 14 mW/cm2. Under these conditions and our experimental values of the measured the reflection coefficient, the typical absolute measurement uncertainties on its magnitude and phase, given by Rohde & Schwarz, were, respectively, 0.0125 to 0.019 in linear (corresponding to 4%-5% of relative uncertainty) and about 2° to 2.5°. A standard calibration was performed at the air waveguide aperture with the manufacturer calibration kit. The calibration plane is shown in Figure 1.

Fig. 1.

Fig. 1

Open in a new tab

Upper image: Open-ended waveguide showing flange with Teflon-filled central aperture. Lower image: Longitudinal cross section of the waveguide.

We noticed that pressing the waveguide aperture directly against the skin made the phase of the reflection coefficient change considerably as a function of the exerted contact pressure. This is mainly because the contact is not flat; when pressed, the skin more or less penetrates inside the waveguide. This configuration where the phase of the reflection coefficient is around ± π makes phase measurements quite unstable, inaccurate and poorly repeatable. As this was unacceptable for our study, we decided to use a quarter-wavelength (at mid-band) Teflon impedance transformer, as was done and explained by Ghodgaonkar et al. [1987]. It has the advantages of providing a flat contact surface between the flange and the waveguide, as well as allowing good sensitivity for phase and magnitude (reduced reflectivity). We inserted a 0.8 mm thick Teflon plug at the aperture of an extra waveguide section so that it fit tightly within the dimensions of the waveguide. A slight pressure against the skin was sufficient to get stable and reliable measurements; the pressure of the Teflon-filled aperture on the skin had no detectable influence on the measured reflection coefficient. Figure 1 shows a diagram of the measurement structure. The measurement reference plane was shifted to the air / Teflon interface.

Experimental procedure

Ten healthy volunteers (one female and nine males, ranging from 29 to 73 years old) participated in this study with the ethical approval of the Temple University Institutional Review Board.

For reasons of convenience with regard to the available instrumentation, we chose an easy-to-access acupuncture point on the forearm, for which the sweat gland density is lower than the palm, as recommended by Ahn et al. [2008]. Instead of measuring one spot, we measured nine spots, one centered on the alleged acupuncture point and eight immediately around it; we measured a rectangular area of about 6 mm × 12 mm, made of 3×3 aperture-sized rectangles, as shown in Figure 2. It was done because the location of acupuncture points is not anatomically defined with a high precision. These nine measurements were averaged for each frequency and each subject.

Fig. 2.

Fig. 2

Open in a new tab

Forearm showing location of acupuncture site and ipsilateral control site. The white line running from the tip of middle finger to axilla is the course of the pericardium meridian.

We chose the so-called PC4 (PeriCardium No. 4) Xi-men point on the pericardium meridian of hand (according to Traditional Chinese Medicine). It is situated on the anterior part of the forearm, at 5 / 12 of the distance between the wrist and the fold of the elbow, starting from the wrist, as shown in Figure 2.

The measurement at the acupuncture site must be compared to a reference, which is obviously in an area excluding any pre-identified acupuncture point. Because the skin is not homogeneous (presence of sweat glands, hair follicles, etc.), one can expect local variations in the reflection coefficient, but also differences when measuring distant parts of the body. That is why choosing only one ‘control’ site, either near the acupuncture point or on a different part of the body, could create differences not necessarily due to the presence of an acupuncture point. Therefore, two controls are needed and they should be compared to each other. A first control is taken near the acupuncture point on the same forearm and an additional one is chosen at the very same site but on the other forearm. These two controls are symmetric to each other on the body, hence comparable, and were expected to have similar properties. The second control is thus a control for the first one. They were both 2.5 cm away from the PC4 acupuncture points at a 45-degree angle with the PC meridian. Nine measurements were also taken for each of these control sites, as shown in Figure 2. Therefore, three sites were measured, with nine spots in each. The same PC4 acupuncture point and the two controls had to be located on both sides of the body; then the side at which the acupuncture point was measured was randomized, as well as the order in which the measurements were carried out.

When necessary, the studied areas were carefully shaved to avoid the effect of hair at the contact interface between skin and waveguide flange. Volunteers had to make contact with the demarcated areas on their skin with the waveguide probe for each of the 27 measured spots; this procedure was controlled by the experimenter.

Probe model and calculation method of dielectric properties

For each frequency, the real and imaginary parts of the complex permittivity were reconstructed from the measured complex reflection coefficient (saved in the form of magnitude and phase) using a suitable model of the aperture of an open-ended rectangular waveguide radiating into a lossy dielectric half-space. We used the model originally developed by Lewin [1951] as extended by Ganchev et al. [1992] to the case of lossy dielectrics. This approximate model is based on the assumption that only the TE10 dominant mode need be taken into account to calculate the aperture admittance. Indeed, in our case, taking into account the higher order modes at the aperture is not necessary, as deduced from the comprehensive analysis conducted by Bois et al. [1999]. It was shown that these higher order modes are only necessary when dealing with low permittivity (e.g., below 5) and low-loss dielectric (e.g., below 1) materials; the contribution of higher order modes actually decreases as the permittivity increases, while the reflection coefficient increases. From Alekseev and Ziskin [2007], it can be inferred that the conditions of the approximation are met in our frequency range in the case of forearm skin. The presence of the Teflon impedance transformer at the aperture instead of air is not problematic since the next higher order reflected mode (TE30 mode) is non-propagating in the Teflon, but evanescent (cut-off frequency above 83 GHz). Moreover, we observed no resonant perturbation in the admittance because they only occur above the TE30 cutoff [Swift, 1969]. Indeed, as in Ghodgaonkar et al. [1987], Teflon was assumed to be lossless and to have a relative permittivity of εT = 2.05. We adapted the model by Ganchev et al. [1992] to take into account the Teflon plug. The normalized aperture admittance Yap is then expressed as:

Yap=2jπabk10∫0a∫0b(b−x)h(y)e−jkrx2+y2x2+y2dxdy

Where j2=-1;k10=εTk02−kc2 is the propagation constant in Teflon of the TE10 mode, k0 is the propagation constant in free space, and kc = π/a is the cutoff wave number; h(y)=K2(a−y)cos(πya)+aπK1sin(πya) with K1=kr2+kc2,K2=kr2−kc2,kr=k0εr and εr = εr’ – j εr” being the complex relative permittivity to be determined.

The model supposes that the material being measured is lossy enough to be considered as infinite in thickness (our case), and reasonably homogeneous. This was shown to be a good approximation for the forearm skin by Alekseev and Ziskin [2007]; so we supposed that the measured skin regions will be homogeneous in our case. The metallic waveguide flange is supposed to be infinite, which is not a problem in practice due to the dielectric losses in the skin.

Beforehand, the reflection coefficient at the measurement reference plane (air/Teflon interface) had been transformed into the normalized aperture admittance of the open-ended waveguide by an appropriate impedance transformation. A solving procedure was implemented using the software Matlab (The MathWorks, Natick, MA, USA) to calculate the dielectric properties for each frequency.

Statistical analyses

We tested the null hypothesis that there were no differences among the means coming from the group that includes the PC4 acupuncture point and the two control groups. Because of the uncertainty in the exact location of the acupuncture point, which made us average the 9 measured spots, a significance level of 0.1 was used in this study. Analyzing the data should take into account individual differences. A one-way within-subject analysis of variance (also called repeated measures ANOVA) was carried out to compare the three sites among all subjects, for both εr’ and εr”. In case of statistically significant differences among the three groups, a subsequent Bonferroni t-test approach was used for multiple comparisons. Before undertaking ANOVA, the normality of each variable for each group (acupuncture site, control site on the same side of the arm, control site on the opposite side) and homoskedasticity were checked to satisfy the Kolmogorov-Smirnov test and Levene's test over the whole frequency range. As specifically required by repeated measures ANOVA, non-sphericity was corrected for using the Huynh–Feldt correction approach.

All analyses and corresponding tests were all conducted for each frequency with the help of Matlab functions and scripts.

RESULTS

We checked that our method was able to detect differences between single measurements made on different spots belonging to different sites. Most of these differences were greater than the measurement uncertainty.

For each of the three sites tested, the 9 measured spots were first averaged as explained earlier, then the average of the corresponding means was calculated across the 10 volunteers; the resulting overall averages of the three groups are shown on Figure 3. The standard deviations were added to provide information on dispersion. Some trends can be noticed mainly in the lower part of the explored frequency range; εr’, the real part of the complex permittivity of the acupuncture group, was found lower than the two controls, while εr”, the imaginary part, was found slightly higher.

Fig. 3.

Fig. 3

Open in a new tab

Dielectric properties of the skin of all subjects at the acupuncture and control sites (for most of the frequencies, the dielectric properties of the controls overlap). Group means and standard deviation error bars are presented. (a) Real part of permittivity. (b) Imaginary part.

Within-subject ANOVA

The p-values resulting from the repeated measures ANOVA relative to εr’ reveal that there is a statistically significant difference among the means of the three groups for frequencies below 61 GHz, for which the p-value is below 0.1; it goes down to 0.07 – 0.08 (but after 61 GHz, 0.1 < p < 0.28). However, the differences observed for εr” were not statistically significant (0.1 < p < 0.3). Because three further multiple comparisons were made regarding εr’ (for frequencies below 61 GHz), the corresponding Bonferroni significance level is 3.3%. The corresponding pair-wise p-values showed that the two control sites were not statistically different for εr’ (p > 0.6), nor was the acupuncture site and the control on the other arm (0.1 < p < 0.2), but the acupuncture site and the control site on the same side of the arm were statistically different (p < 0.033).

DISCUSSION

The aim of the study was to see if the skin at an acupuncture point could exhibit different dielectric properties at millimeter wavelengths, which could be a biophysical basis for their use in millimeter wave therapy. Our study suggests that the dielectric properties of the PC4 acupuncture site are different from surrounding skin in the 50-61 GHz range, actually the real part of the complex permittivity, εr’.

The open-ended rectangular waveguide technique has been used to measure the dielectric properties of materials, especially biological tissues and human skin, in the millimeter wave range [Hey-Shipton et al., 1982; Ghodgaonkar et al., 1987; Alekseev and Ziskin, 2007]. The state-of-the-art measurement technique used in this study provided reliable and stable measurements of the reflection coefficient on the skin surface, notably regarding its phase. The method used to calculate the dielectric properties of the skin from the measurement was carefully examined and chosen.

In this frequency range, the penetration depth is very shallow and the dielectric properties depend mostly on water content. The lack of statistical significance after 61 GHz may be related to the smaller penetration of the waves into the skin at these frequencies. There could be a particularity in the structure of the skin at acupuncture points that was detected at the explored millimeter wave frequencies. The statistically lower value of the permittivity εr’ at the acupuncture site could be interpreted in terms of the different water content or the distribution of water molecules around collagen fibers within the skin; the potentially higher value of losses (as evaluated by εr”) might correspond to a slightly better absorption of millimeter waves at the acupuncture site, but these considerations are only speculations at this point. If a singularity exists at these specific sites on the body, it seems likely to be below the skin in the subcutaneous connective tissue, according to previous research [Langevin et al., 2002; Langevin and Yandow, 2002; Ahn et al., 2005]. Actually, as pointed out by Ahn et al. [2008], the electrical difference seems more likely to be due to a connective tissue path (a so-called ‘meridian’) and detectable at low frequencies than on a specific spot (a so-called ‘acupuncture point’) [Ahn et al., 2005]. However, our results are consistent with the ones obtained by researchers looking at acupuncture points and meridians at optical frequencies [Choi et al., 2003; Yang et al., 2007]. Nevertheless, red or near infrared light can penetrate deeper into the body than millimeter waves [Anderson and Parrish, 1981; Barun et al., 2007].

Undeniably this study has a few limitations. For convenience and practicality reasons, we opted for a proportional measurement as far as locating the acupuncture point is concerned; admittedly, this method is prone to inaccuracies. Instead of locating the point, a scanning bench set-up and scanning methodology around a pre-identified area would be better because it limits the possibility that the site is inappropriately identified and allows for potential local variations to be detected. Studies that predetermined the acupuncture site prior to electrical measurement do not usually find a positive association, as emphasized by Ahn et al. [2008]. Moreover, the size of an acupuncture ‘point’ is still uncertain and unclear, especially at millimeter wave frequencies. It could be smaller or larger than the aperture of the waveguide we used. If it can be localized with a good scanning technique, then an approximate size can be given. Furthermore, the homogeneous skin model used to calculate the dielectric properties only gives us an effective, macroscopic complex permittivity – a sort of average value over a small skin volume around the probe aperture – and does not account for very local variations.

Moreover, our study was not strictly double-blinded, but the presence of two controls prevented it; the subjects knew that one control site was measured on one forearm, and that the symmetrical control and acupuncture site was measured on the other forearm. However, the very good measurement method in which pressure had no influence on the outcome, as well as the computerized saving of the data, reduced potential experimental bias from the experimenter and participant. It was almost as if the experiment was single-blinded. Potential confounders could be the relative presence of hair follicles on the different sites for the different subjects; even if the corresponding areas were shaved, the follicles remain in the skin. Another confounding factor could be the different superficial temperature of the skin at the acupuncture site from the controls due to a different blood supply.

We should be cautious about the conclusion of the study due to some limitations. Our noteworthy and non-expected results would deserve a more comprehensive, larger-scale study with more subjects. It would be preferable to set up an improved procedure, which could involve a scanning methodology or a blinding protocol, as well as measuring at lower GHz frequencies and testing other acupuncture points.

CONCLUSION

At the start of this study it was not identified if there was any biophysical basis for the practice of applying MMWT at acupuncture points. However, this study tends to show that the permittivity of the PC4 acupuncture site is different from that of surrounding non-acupuncture sites in the 50 - 61 GHz range. Because of this difference in permittivity, it is possible that that the response of the body to millimeter waves differs if this acupuncture site is used as an exposure site, rather than surrounding skin. But as we cannot directly conclude that on the basis of this study, further research is called for to determine whether the application of MMWT at acupuncture sites is truly more efficacious than at non-acupuncture sites.

ACKOWLEDGEMENTS

The authors wish to thank the following persons: all volunteers for their collaboration in the study; Dr. Alekseev for helpful discussions; Drs. Helene Langevin and Andrew Ahn for useful discussions about testing acupuncture points; Drs. Xiao-Hong Chen and Wei Xu for their help in locating acupuncture spots; Drs. Karl Bois, Sergiy Kharkovsky and Reza Zoughi for useful exchanges about the open-ended waveguide method; Dr. Deepak Ghodgaonkar for his PhD thesis and useful information; and Dr. Kwang-Sup Soh for providing some of his papers. In addition, the laboratory thanks Rohde & Schwarz, Agilent Technologies, and Anritsu for providing equipment and technical support during this study.

Grant sponsor: NIH NCCAM, Grant Number P01-AT002025.

REFERENCES

  1. Ahn AC, Wu J, Badger GJ, Hammerschlag R, Langevin H. Electrical impedance along connective tissue planes associated with acupuncture meridians. BMC Complem Altern Med. 2005;5(10):1–9. doi: 10.1186/1472-6882-5-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ahn AC, Colbert AP, Anderson BJ, Martinsen OG, Hammerschlag R, Cina S, Wayne PM, Langevin H. Electrical properties of acupuncture points and meridians: A systematic review. Bioelectromagnetics. 2008;29:245–256. doi: 10.1002/bem.20403. [DOI] [PubMed] [Google Scholar]
  3. Alekseev SI, Ziskin MC. Human skin permittivity determined by millimeter wave reflection measurements. Bioelectromagnetics. 2007;28(5):331–339. doi: 10.1002/bem.20308. [DOI] [PubMed] [Google Scholar]
  4. Alekseev SI, Radzievsky AA, Logani MK, Ziskin MC. Millimeter wave dosimetry of human skin. Bioelectromagnetics. 2008;29:65–70. doi: 10.1002/bem.20363. [DOI] [PubMed] [Google Scholar]
  5. Anderson RR, Parrish JA. The optics of human skin. J Investig Dermatol. 1981;77:13–19. doi: 10.1111/1523-1747.ep12479191. [DOI] [PubMed] [Google Scholar]
  6. Barun VV, Ivanov AP, Volotovskaya AV, Ulashchik VS. Absorption spectra and light penetration depth of normal and pathologically altered human skin. J Applied Spectroscopy. 2007;74(3):430–439. [Google Scholar]
  7. Bois KJ, Benally AD, Zoughi R. Multimode solution for the reflection properties of an open-ended rectangular waveguide radiating into a dielectric half-space: The forward and inverse problems. IEEE Trans Instrum Meas. 1999;48(6):1131–1140. [Google Scholar]
  8. Chernavsky DS, Karp VP, Rodshtat IV. On the neurophysiological mechanism of EHF punctural therapy. In: Deviatkov ND, Betskii OV, editors. Biological aspects of low intensity millimeter waves. Moscow University; Moscow, Russia: 1994. pp. 164–206. [Google Scholar]
  9. Choi C, Soh KS, Lee SM, Yoon G. Study of propagation of light along an acupuncture meridian. J of the Optical Society of Korea. 2003;7(4):244–247. [Google Scholar]
  10. Falk CX, Birch S, Avants SK, Tsau Y, Margolin A. Preliminary results of a new method for locating auricular acupuncture points. Acupunct Electrother Res. 2000;25(3-4):165–177. doi: 10.3727/036012900816356136. [DOI] [PubMed] [Google Scholar]
  11. Feng IC, Ifrim M. Further study on the anatomical, histological and biochemical bases underlying clinical acupuncture effectiveness. J Chin Med. 2004;15(2):69–78. [Google Scholar]
  12. Ganchev SI, Bakhtiari S, Zoughi R. A novel numerical technique for dielectric measurement of generally lossy dielectrics. IEEE Trans Instrum Meas. 1992;41(3):361–365. [Google Scholar]
  13. Ghodgaonkar DK. PhD dissertation. Department of Electrical Engineering, University of Utah, Salt Lake City, UT, USA; 1987. Measurement of complex permittivities of biological materials and human skin in vivo in the frequency band 26.5 – 60 GHz. [Google Scholar]
  14. Hey-Shipton GL, Matthews PA, McStay J. The complex permittivity of human tissue at microwave frequencies. Phys Med Biol. 1982;27(8):1067–1071. doi: 10.1088/0031-9155/27/8/008. [DOI] [PubMed] [Google Scholar]
  15. Hyvarinen J, Karlsson M. Low-resistance skin points that may coincide with acupuncture loci. Med Biol. 1977;55(2):88–94. [PubMed] [Google Scholar]
  16. Langevin HM, Yandow JA. Relationship of acupuncture points and meridians to connective tissue planes. Anat Res. 2002;269:257–265. doi: 10.1002/ar.10185. [DOI] [PubMed] [Google Scholar]
  17. Langevin HM, Churchill DL, Wu J, Badger GJ, Yandow JA, Fox JR, Krag MH. Evidence of connective tissue involvement in acupuncture. FASEB J. 2002;16:872–874. doi: 10.1096/fj.01-0925fje. [DOI] [PubMed] [Google Scholar]
  18. Lewin L. Advanced theory of waveguides. Illiffe and Sons; London, UK: 1951. [Google Scholar]
  19. Li AH, Zhang JM, Xie YK. Human acupuncture points mapped in rats are associated with excitable muscle/skin-nerve complexes with enriched nerve endings. Brain Research. 2004;1012:154–159. doi: 10.1016/j.brainres.2004.04.009. [DOI] [PubMed] [Google Scholar]
  20. Pakhomov AG, Prol HK, Mathur SP, Akyel Y, Campbell CBG. Search for frequency-specific effects of millimeter-wave radiation on isolated nerve function. Bioelectromagnetics. 1997;18:324–334. doi: 10.1002/(sici)1521-186x(1997)18:4<324::aid-bem5>3.0.co;2-4. [DOI] [PubMed] [Google Scholar]
  21. Pakhomov AG, Murphy MR. Low-intensity millimeter waves as a novel therapeutic modality. IEEE Trans on Plasma Science. 2000;28(1):34–40. [Google Scholar]
  22. Radzievsky AA, Rojavin MA, Cowan A, Alekseev SI, Radzievsky AA, Jr, Ziskin MC. Peripheral neural system involvement in hypoalgesic effect of electromagnetic millimeter waves. Life Sci. 2001;68:1143–1151. doi: 10.1016/s0024-3205(00)01016-x. [DOI] [PubMed] [Google Scholar]
  23. Radzievsky AA, Gordiienko OV, Cowan A, Alekseev SI, Ziskin MC. Millimeter-wave-induced hypoalgesia in mice: Dependence on type of experimental pain. IEEE Trans on Plasma Science. 2004;32(4):1634–1643. [Google Scholar]
  24. Reichmanis M, Marino AA, Becker RO. Electrical correlates of acupuncture points. IEEE Trans Biomed Eng. 1975;22(6):533–535. doi: 10.1109/tbme.1975.324477. [DOI] [PubMed] [Google Scholar]
  25. Reichmanis M, Marino AA, Becker RO. DC skin conductance variation at acupuncture loci. Am J Chin Med. 1976;4(1):69–72. doi: 10.1142/s0192415x7600010x. [DOI] [PubMed] [Google Scholar]
  26. Rojavin M, Ziskin MC. Medical application of millimeter waves. Q J Med. 1998;91:57–66. doi: 10.1093/qjmed/91.1.57. [DOI] [PubMed] [Google Scholar]
  27. Swift C. Admittance of a waveguide-fed aperture loaded with a dielectric plug. IEEE Trans Antennas Propag. 1969;17:356–359. [Google Scholar]
  28. Usichenko YI, Edinger H, Gizhko VV, Lehmann C, Wendt M, Feyerherd F. Low-intensity electromagnetic millimeter waves for pain therapy. Evidence-based Complementary and Alternative Medicine. 2006;3:201–207. doi: 10.1093/ecam/nel012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Wick F, Wick N, Wick MC. Morphological analysis of human acupuncture points through immunohistochemistry. Am J Phys Med Rehabil. 2007;86(1):7–11. doi: 10.1097/01.phm.0000250564.88013.89. [DOI] [PubMed] [Google Scholar]
  30. Yang HQ, Xie SS, Liu SH, Li H, Guo ZY. Differences in optical properties between human meridian and non-meridian. Am J Chin Med. 2007;35(5):743–52. doi: 10.1142/S0192415X07005235. [DOI] [PubMed] [Google Scholar]
  31. Zhao ZQ. Neural mechanism underlying acupuncture analgesia. Progress in Neurobiology. 2008;85:355–375. doi: 10.1016/j.pneurobio.2008.05.004. [DOI] [PubMed] [Google Scholar]

RESOURCES