22 May 2018 Ultrasonic standing wave preparation of a liquid cell for glucose measurements in urine by midinfrared spectroscopy and potential application to smart toilets
Author Affiliations +
Smart toilets could be used to monitor different components of urine in daily life for early detection of lifestyle-related diseases and prompt provision of treatment. For analysis of biological samples such as urine by midinfrared spectroscopy, thin-film samples like liquid cells are needed because of the strong absorption of midinfrared light by water. Conventional liquid cells or fixed cells are prepared based on the liquid membrane method and solution technique, but these are not quantitative and are difficult to set up and clean. We generated an ultrasonic standing wave reflection plane in a sample and produced an ultrasonic liquid cell. In this cell, the thickness of the optical path length was adjustable, as in the conventional method. The reflection plane could be generated at an arbitrary depth and internal reflected light could be detected by changing the frequency of the ultrasonic wave. We could generate refractive index boundaries using the density difference created by the ultrasonic standing wave. Creation of the reflection plane in the sample was confirmed by optical coherence tomography. Using the proposed method and midinfrared spectroscopy, we discriminated between normal urine samples spiked with glucose at different concentrations and obtained a high correlation coefficient.
Yamamoto, Kawashima, Kitazaki, Mori, Kang, Nishiyama, Wada, and Ishimaru: Ultrasonic standing wave preparation of a liquid cell for glucose measurements in urine by midinfrared spectroscopy and potential application to smart toilets



Early treatment is important in the prevention of lifestyle-related diseases, such as diabetes and gout.1 Smart toilets that could measure glucose and protein in urine in daily life are of interest in this area. To date, smart toilets can only control cleaning and measure the volume of urine using flow rate sensors, motion detectors, and other technology.2 Currently, there is no sensor that can be installed in a toilet and measure different components of interest in urine. If glucose or protein is present in urine, it is possible that the person has diabetes or chronic kidney disease.3,4 Early detection of these diseases by smart toilets in daily life could prompt an individual to seek treatment at an early stage of the disease.5,6 Many approaches have been investigated for smart toilets and noninvasive blood glucose sensors, with most using near-infrared spectroscopy7,8 and wavelengths from 800 to 1500 nm. Near-infrared light can pass through a sample more easily than midinfrared light because it is absorbed less by water.9 Therefore, near-infrared spectroscopy can be used for noninvasive analysis of biological samples. However, the near-infrared absorption of glucose is weak because the absorption peak in the near-infrared region appears as an overtone and combination band.10 Furthermore, it is difficult to identify the absorption peak of glucose because there are also absorption peaks for water, proteins, and hemoglobin in the near-infrared region.11,12

Smart toilets using midinfrared spectroscopy could be used to detect the fundamental vibration of glucose. We have already proposed a small (bean-sized) midinfrared Fourier spectroscopic imager.13 However, because midinfrared light is absorbed strongly by water, this imager requires thin-film samples with thicknesses of <100  μm for midinfrared spectroscopy detection of transmitted or reflected light. Conventional liquid cells and fixed cells prepared based on the liquid membrane method and solution techniques are used for optical transmission measurements.14 Unfortunately, these cells are not quantitative and are difficult to set up and clean.15 Another option is the attenuated total reflectance method,1617.18 which uses an evanescent wave that can travel several micrometers into a sample. However, the light must be reflected many times in the attenuated total reflectance prism to meet the required optical path length, and light intensity is an issue because increased reflection leads to increased absorption by the prism and samples.

In this paper, we describe development of an ultrasonic liquid cell, with a reflection plane generated inside the sample by an ultrasonic standing wave, for midinfrared spectroscopy and application to smart toilets [Fig. 1(a)]. We used optical coherence tomography (OCT) to investigate the reflection plane of the ultrasonic standing wave in the sample and applied our method to normal urine samples spiked with different concentrations of glucose. The urine used in the study was collected from a 24-year-old adult male. Additionally, we determined the correlation coefficient for measurement of the glucose concentration to evaluate the feasibility of this method for quantitative measurement of glucose concentrations and the realization of smart toilets. Glucose is typically excreted from blood into urine when the blood glucose level reaches around 180  mg/dL.19 However, because this level can vary, we instead used the level (50  mg/dL) for a positive result with a urine test strip. For the realization of smart toilets, we believe that a glucose target concentration of 50  mg/dL and measurement accuracy of 30  mg/dL will be relevant.

Fig. 1

(a) Schematic diagram of a smart toilet and (b) the liquid cell with variable optical path length.



Principles of the Ultrasonic Liquid Cell and Verification Experiments

If a sample is homogeneous, some incident light will be detected as surface reflected light. However, light absorbed within the sample cannot be detected. Therefore, in this method, an ultrasonic standing wave was used to generate refractive index boundaries inside the sample, and internal reflected light was detected at an arbitrary depth [Fig. 1(b)] Because the ultrasonic wave is a compressional wave, it propagates while generating a density difference within the sample. The refractive index distribution is also stabilized by the ultrasonic standing wave. The incident light is reflected at the node of the ultrasonic standing wave where the refractive index differences are maximized, and it is detected as internal reflected light. The position of the node of the ultrasonic standing wave depends on the frequency of the ultrasonic wave. By manipulating the frequency of the ultrasonic standing wave, we could obtain an arbitrary optical path length in the depth direction.

In this study, the ultrasonic standing wave was generated inside a container fabricated of BaF2 optical windows with high transmittance of midinfrared light, and the internal reflection plane was confirmed using OCT (IVS-2000, Santec, Komaki, Japan). Figure 2(a) shows the optical system. To visualize the generated reflective surface, pure water containing red fluorescent polymer microspheres (36-3, Thermo Fisher Scientific, Waltham, Massachusetts) was used as a sample. An ultrasonic transducer (PN-10C10N, Japan Probe Co. Ltd., Yokohama, Japan) was attached behind the target, and vibrated at a frequency of 10 MHz and a voltage of 10 V to generate an ultrasonic standing wave inside the target. The OCT image confirmed that the particles in the water aggregated at the node of the ultrasonic standing wave and formed lines [Fig. 2(c)]. The particles near the boundaries with the optical windows were not trapped by the standing wave and did not aggregate. Because incident light is reflected by a node with a large refractive index difference, the incident light is considered to be reflected 0.05 mm from the wall surface. In this case, the optical path length is the sum of that of the incident light and the reflected light from the reflection plane. We selected a reflectance depth of 0.05 mm, because it provided an optical path length of about 100  μm, which is suitable for midinfrared spectroscopy. To generate a reflection plane at this position, a frequency of 10 MHz is required. Even though the sample is homogeneous and internal reflected light cannot usually be detected, it could be detected from node positions by generating a refractive index difference with an ultrasonic standing wave.

Fig. 2

A reflection plane generated by an ultrasonic standing wave is detected by OCT. (a) Experimental setup, (b) a detailed image of the container, and (c) the results of an experiment.



Measurement of Glucose in Normal Urine by Midinfrared Spectroscopy

We constructed an optical system for detecting internal reflected light in pure water [Fig. 3(a)]. We used a small graphite light source (EK8620, Helioworks, Santa Rosa, California) with an attached setup for Kohler illumination. To shorten the optical path length, the incident light from the optical system entered through the wall of the target at an angle of about 45 deg. The ultrasonic vibrator was placed on the bottom of the target. We detected internal reflected light using the two-dimensional Fourier spectroscopic imager.20 This light was reflected from the reflection plane, which was generated 0.05 mm from the surface by the ultrasonic standing wave using the wall surface as the reflective material. Next, we compared the relative intensities before and after generation of the ultrasonic wave to confirm detection of internal reflected light by our proposed liquid cell. The relative intensities were evaluated for areas of five pixels by five pixels [Fig. 3(b)]. The relative intensity from 8 to 14  μm after ultrasonic vibration was about 25% more than that before ultrasonic vibration. These results confirmed that internal reflected light was detected by our method. The refractive index of the optical window material (BaF2) is 1.414 at 9  μm.21 Thus, the reflectance after vibration was calculated as 8% using Fresnel equations. 22 Additionally, using the 25% increase in the relative intensity, the reflectance before vibration was calculated as 2% (25% of 8%). Consequently, the refractive index of water was calculated as 1.1 using the reflectance formula,23 and the refractive index difference was around 0.3. This value is equivalent to the refractive index when light is reflected on the reflection plane. In future work, we will verify this and determine the theoretical relationship between the reflected light intensity and the ultrasonic standing wave.

Fig. 3

(a) The optical system for the experiment and (b) comparison of the intensities before and after ultrasonic vibration.


Next, a solution of glucose in urine was poured into the container and we measured the glucose absorbance with the optical system using the internal reflected light [Fig. 3(a)]. We spiked the urine with glucose at three concentrations (50, 100, and 200  mg/dL). These concentrations were selected because they are around the level that is considered a positive result for urine tested with a glucose test strip (100  mg/dL). The relative intensity of the urine after ultrasonic vibration was used as a reference for calculating the absorbance of glucose in urine. We repeated the measurement 30 times and calculated the average. Absorption peaks for glucose were observed at 9.25 and 9.65  μm, respectively [Fig. 4(a)]. The peak at 9.25  μm was not obstructed by other components in the urine [Fig. 4(b)], and we obtained a high correlation coefficient (0.91) for measurement of the glucose concentration in urine using this peak. Therefore, this method is feasible for quantitative measurements of glucose concentrations in urine without a complex setup.

Fig. 4

(a) Absorbance of glucose in urine (n=30) and (b) verification of the measurement accuracy at 9.25  μm.


However, the 3σ value calculated from the standard deviation for the 30 measurements at 9.25  μm was about 0.015, and the reproducibility of this method cannot be guaranteed. This could be caused by absorption of midinfrared light by water in the atmosphere, or by fluctuation of the output from the light source. Our proposed method of one-shot Fourier spectroscopy could overcome this problem, because it has high time resolution.23



Our liquid cell, generated by an ultrasonic standing wave in a sample, could be used for realization of smart toilets using midinfrared spectroscopy. In future work, we will study the optimum shape of this liquid cell for installation into smart toilets. Additionally, the stability and repeatability of the liquid cell for urine glucose and protein measurements will be verified by developing a one-shot Fourier spectroscopic imager for use in the midinfrared region.


We declare no conflict of interest.


We thank the Japan Agency for Medical Research and Development. We thank Gabrielle David, PhD, from Edanz Group ( www.edanzediting.com/ac) for editing a draft of this letter.


1. J. A. M. Bispo et al., “Correlating the amount of urea, creatinine, and glucose in urine from patients with diabetes mellitus and hypertension with the risk of developing renal lesions by means of Raman spectroscopy and principal component analysis,” J. Biomed. Opt. 18(8), 087004 (2013).JBOPFO1083-3668 https://doi.org/10.1117/1.JBO.18.8.087004 Google Scholar

2. D. R. Hall et al., “Smart flush toilet system,” U.S. Patent No. 0254060 A1 (2017). Google Scholar

3. C. Wald et al., “Diagnostics: a flow of information,” DHR Int. J. Biomed. Life Sci. 551, S48–S50 (2017). Google Scholar

4. D. J. F. Rowe et al., “Microalbuminuria in diabetes mellitus: review and recommendations for the measurement of albumin in urine,” Ann. Clin. Biochem. Int. J. Lab. Med. 27(4), 297–312 (1990). Google Scholar

5. S. Saito et al., “Method and apparatus for detecting urinary constituents,” U.S. Patent No. 5073500 (1991). Google Scholar

6. W. R. Premasiri, R. H. Clarke and M. E. Womble, “Urine analysis by laser Raman spectroscopy,” Lasers Surg. Med. 28(4), 330–334 (2001).LSMEDI0196-8092 https://doi.org/10.1002/(ISSN)1096-9101 Google Scholar

7. U. A. Müller et al., “Non-invasive blood glucose monitoring by means of near infrared spectroscopy: methods for improving the reliability of the calibration models,” Int. J. Artif. Organs 20(5), 285–290 (1997).IJAODS0391-3988 https://doi.org/10.1177/039139889702000509 Google Scholar

8. H. M. Heise et al., “Noninvasive blood glucose sensors based on near-infrared spectroscopy,” Artif. Organs 18(6), 439–447 (1994).ARORD70160-564X https://doi.org/10.1111/aor.1994.18.issue-6 Google Scholar

9. H. Büning-Pfaue et al., “Analysis of water in food by near infrared spectroscopy,” Food Chem. 82(1), 107–115 (2003). https://doi.org/10.1016/S0308-8146(02)00583-6 Google Scholar

10. V. A. Corro-Herrera et al., “In-situ monitoring of Saccharomyces cerevisiae ITV01 bioethanol process using near-infrared spectroscopy NIRS and chemometrics,” Biotechnol. Prog. 32(2), 510–517 (2016). https://doi.org/10.1002/btpr.2222 Google Scholar

11. M. Pleitez et al., “Infrared spectroscopic analysis of human interstitial fluid in vitro and in vivo using FT-IR spectroscopy and pulsed quantum cascade lasers (QCL): establishing a new approach to non invasive glucose measurement,” Spectrochim. Acta 85(1), 61–65 (2012).SPACA50038-6987 https://doi.org/10.1016/j.saa.2011.09.007 Google Scholar

12. A. Schwaighofer et al., “Quantum cascade lasers (QCLs) in biomedical spectroscopy,” Chem. Soc. Rev. 46(19), 5903–5924 (2017).CSRVBR0306-0012 https://doi.org/10.1039/C7CS00403F Google Scholar

13. S. Sato et al., “Ultra-miniature one-shot Fourier-spectroscopic tomography,” Opt. Eng. 55(2), 025106 (2016). https://doi.org/10.1117/1.OE.55.2.025106 Google Scholar

14. S. Liakat et al., “In vitro measurements of physiological glucose concentrations in biological fluids using mid-infrared light,” Biomed. Opt. Express 4(7), 1083–1090 (2013).BOEICL2156-7085 https://doi.org/10.1364/BOE.4.001083 Google Scholar

15. Shimadzu Corporation, “Measurement methods for liquid samples,” 2017,  https://www.shimadzu.com/an/ftir/support/ftirtalk/talk9/intro.html (1 December 2017). Google Scholar

16. H. M. Heise et al., “Multivariate calibration for the determination of analytes in urine using mid-infrared attenuated total reflection spectroscopy,” Appl. Spectrosc. 55(4), 434–443 (2001).APSPA40003-7028 https://doi.org/10.1366/0003702011951948 Google Scholar

17. G. Budinova, J. Salva and K. Volka, “Application of molecular spectroscopy in the mid-infrared region to the determination of glucose and cholesterol in whole blood and in blood serum,” Appl. Spectrosc. 51(5), 631–635 (1997).APSPA40003-7028 https://doi.org/10.1366/0003702971941034 Google Scholar

18. S. Kino et al., “Hollow optical-fiber based infrared spectroscopy for measurement of blood glucose level by using multi-reflection prism,” Biomed. Opt. Express 7(2), 701–708 (2016).BOEICL2156-7085 https://doi.org/10.1364/BOE.7.000701 Google Scholar

19. R. Guptal et al., “Diabetes mellitus: the pandemic of 21st century!” Asian J. Exp. Biol. Sci. 23(1), 261–268 (2009). Google Scholar

20. Y. Inoue et al., “Variable phase-contrast fluorescence spectrometry for fluorescently stained cells,” Appl. Phys. Lett. 89, 121103 (2006).APPLAB0003-6951 https://doi.org/10.1063/1.2356312 Google Scholar

21. “Refractive index of barium fluoride (BaF2), aerosol refractive index archive,” Earth Observation Data Group, 2018,  http://eodg.atm.ox.ac.uk/ARIA/data?Salts/Barium_Fluoride_(Querry_1987)/BaF2_Querry_1987.ri (20 March 2018). Google Scholar

22. B. E. A. Saleh et al., Fundamentals of Photonics, Wiley-Interscience, New York (1991). Google Scholar

23. V. Ivanovski et al., “Polarized IR reflectance spectra of the monoclinic single crystal K2Ni(SO4)2·6H2O: dispersion analysis, dielectric and optical properties,” Spectrochim. Acta Part A 69(2), 629–641 (2008). https://doi.org/10.1016/j.saa.2007.05.014 Google Scholar

24. A. Ishida et al., “Quantitative measurement of biological substances in daily-life environment with the little-finger-size one-shot spectroscopic tomography,” Proc. SPIE 8951, 89510Y (2014).PSISDG0277-786X https://doi.org/10.1117/12.2038836 Google Scholar

© The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.
Naoyuki Yamamoto, Naoyuki Yamamoto, Natsumi Kawashima, Natsumi Kawashima, Tomoya Kitazaki, Tomoya Kitazaki, Keita Mori, Keita Mori, Hanyue Kang, Hanyue Kang, Akira Nishiyama, Akira Nishiyama, Kenji Wada, Kenji Wada, Ichiro Ishimaru, Ichiro Ishimaru, } "Ultrasonic standing wave preparation of a liquid cell for glucose measurements in urine by midinfrared spectroscopy and potential application to smart toilets," Journal of Biomedical Optics 23(5), 050503 (22 May 2018). https://doi.org/10.1117/1.JBO.23.5.050503 . Submission: Received: 11 December 2017; Accepted: 1 May 2018
Received: 11 December 2017; Accepted: 1 May 2018; Published: 22 May 2018

Back to Top