The accurate assessment of dehydration is crucial in many diverse clinical applications. Currently used methods for assessing dehydration rely on either skin pinch tests or analysis of urine. Therefore, therefore is a need for wearable non-invasive devices for continuous dehydration monitoring. This paper presents a novel sensor design for the monitoring of dehydration levels by the use of chipless microwave resonators. The sensor design incorporates a metallic layer beyond the tag sensor itself, resulting in an isolation of the dehydration sensing system from conflicting ambient signals that provides a targeted sensing system to the tissue itself with reduced interference. The sensitivity of the sensor is high, with a ~100 KHz shift for a 1% change in dehydration).
Article
Non-Invasive Microwave Sensor Design for Real-Time Continuous Dehydration Monitoring
https://doi.org/10.21203/rs.3.rs-3001466/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Microwave sensor
dehydration
non-invasive
chipless sensor
frequency deviation
Dehydration occurs when the human body loses too much water and other fluids that it needs to work normally. Dehydration is common in infants due to their inability to express thirst, in athletes due undergoing intense exercise in the elderly because of reduced function of their brains thirst center [1], especially in people suffering from degenerative brain dementia diseases such as Alzheimer [2], [3]. Moreover, the increased incidence and severity of heatwaves induced by climate change also increases the risk of dehydration in a significant proportion of the World’s population. Severe dehydration can be life threatening, resulting in low blood pressure, fainting, to kidney, liver, and brain damage, especially as brain function is impaired, leading to confusion [4][5].
The most widely used test for dehydration is based on a visual skin inspection known as turgor test [6]. This simple test involves a skin pinch or press, with a longer time to return to normal shape indicating dehydration. This test is purely qualitative, often inaccurate, and requires some training to undertake [7]. Further quantitative testing may be required that include laboratory-based measurements of urine conductivity and osmolality [8], or blood plasma osmolality [9]. Other methods include isotope dilution [10], and bioelectrical impedance spectroscopy [11]. All of these quantitative tests take time and very sophisticated facilities are staffed by trained professionals. [12]. Dehydration can be easily prevented or treated by either ingesting fluids or via a saline drip. Therefore, there is a clear need to develop non-invasive methods to assess dehydration that provide quantitative results in real-time and at low cost.
Several current dehydration monitoring technologies under development involve real-time analysis of sweat using either chemical sensors or electrical conductance spectroscopy systems [13]–[15]. There are significant disadvantages to this technology that include: 1) sensor saturation after a relatively short period of time, requiring sensor replacement. 2) severe dehydration results in sweat reduction. 3) dehydration is not always associated with sweating and subjects may simply be not ingesting enough fluids. 4) collection of sweat is a major problem associated with all the sweat related biosensors in general [16], [17]. 5) humidity sensing near skin as a surrogate marker for dehydration is extremely susceptible to the ambient environmental conditions [18]. Other technologies are also under development such as digital image processing [19] and wide-band microwave spectroscopy [20], [21]. Although useful, this latter technology is not portable or wearable and requires sophisticated processing.
Microwave split ring resonator (SRR) based sensors may be ideally suited to development as dehydration sensors. SRRs possess high quality factors, are extremely low cost, are intrinsically non-invasive, and can be used for a wide variety of applications [22]–[26] and miniaturized for wearable electronics [27]–[34]. In this paper we present a new SRR-based sensor for the real-time continuous monitoring of dehydration. In this design, the earlobe is considered to be an ideal site for sensor placement as it can be sandwiched between the reader and the tag and therefore variations in its permittivity and conductivity due to changes in the interstitial fluid water content reflect the dehydration level via frequency and amplitude changes in SRR transmission response. This arrangement presents even higher sensitivity than the traditional chipless sensors. In order to reduce environmental signal contamination, a metallic layer has been added to the external surface of the tag resulting in a reduction of additional undesirable sensor errors.
a. Sensor structure
Based on their high sensitivity, zero power consumption, moderate to high quality factor, and the non-invasive nature of the sensor, chipless microwave split ring resonators are excellent candidates for biomedical sensing applications and wearable electronics. The presented structure in this work is a gap coupled transmission line (GCTL) as the reader electromagnetically is coupled to a chipless split ring resonator (tag). The sensor is energized through electromagnetic coupling to a relatively wide-band spectrum provided by the GCTL structure. Research has shown that defecting the ground plane’s conductor of the reader increases the coupling strength between the tag and the reader potentially leading to an increase in the maximum achievable distance between them [31]. However, the sensing mechanism presented herein benefits from this design with some further modifications. As shown in Fig.1, the reader part of the sensor including the GCTL, and the defected ground plane is placed at one side of the tissue under-the test (TUT) and the tag is located on the other side. On the other word, the TUT is sandwiched between the reader and the tag which is different from the traditional sensing mechanisms in which the material under-test located on top of the tag beyond the reader. This modification improves the functionality of the sensor in a high permittivity environment such as the human body that possesses a high conductivity. In addition, the strong coupling between the tag and the reader, due to the impact of the defected ground, results in an enhanced sensitivity that is critical in biomedical applications. Since the majore measured output of these sensors is their resonance frequency, the sensitivity is defined as the shift in their resonance frequency due to change in the permittivity of the material under the test. To obtain further insights to this improved sensitivity of the proposed structure, a field distribution simulation in High-Frequency Structure Simulator (HFSS) is presented in Fig.1(d). The strength of the electromagnetic field in the volume between the reader and the tag is stronger than that beyond the tag and therefore a higher sensitivity is expected.
b. Frequency design
The resonance frequency of the split ring resonator is a function of their effective length and the effective permittivity of their ambient and is defined as follows.
where c is the speed of light, l is the length of the resonator and εeff is the effective permittivity of the resonator’s ambient. To select an optimum resonance frequency for the sensor, it is worth considering some specific characteristics. The permittivity and conductivity of materials are known to vary with frequency, due to different relaxation times. Thus, selecting the resonance frequency of the split ring resonator carefully can result in a higher frequency shift (i.e., sensitivity) for the same concentration variation of materials. This can lead to an improvement in the amplitude response of the sensor, which is directly related to the overall reliability of the system. The dimensions of the various components of the designed sensor are presented in Figure 1 (a)-(c).
The earlobe is chosen as a potential sensor site as it is a relatively simple tissue devoid of bones, ligaments, and major blood vessels (Fig.2(a)). Moreover, the earlobe is of a similar size in most people with general applicability to the majority of the population. The presence of an even distribution of smaller blood vessels makes the earlobe a relatively homogenous and ideal substrate for this particular sensor application [35], [36]. The average thickness of human earlobe is about 3.4mm [34]. Although there is inter-individual variability in thickness of this parameter, it is not very critical for this work as the variability falls in a limited range which supported by the proposed sensor. The different tissues layers within the earlobe and their respective dielectric properties are outlined in Table I. Based on the data provided in Table I, the model (Fig.2(b)) is built for the simulation.
Table I. Earlobe tissue type typical thickness and dielectric properties [35]-[38].
|
|
Thickness |
εr @900MHz |
Conductivity (S/m) |
|
Skin |
0.68 mm |
30 |
0.7 |
|
Fat |
0.78 mm |
14 |
0.1 |
|
Blood |
0.48 mm |
71 |
1.1 |
|
Fat |
0.78 mm |
14 |
0.1 |
|
Skin |
0.68 mm |
30 |
0.7 |
Based on studies that address the aforementioned tissues and materials [37]–[40], and to maintain reasonable permitivities and conductivities while keeping the size of the sensor small enough for earlobe sensing, resonance frequencies in the range of 900MHz to 2GHz are appropriate. Field analysis of the designed resonator over the modelled earlobe tissue is presented in Fig.3. Notably, the electromagnetic field reliability penetrates reliably this model that results in a high quality factor transmission response (Fig. 3(b)). Both peak and notch frequencies could be defined as the output of the sensor, but according to the higher quality factor of the notch frequency (which is the transmission zero of the sensor), it is considered as the sensor’s output. Possessing a high quality factor results in higher resolution in the measurement which is a highly desirable feature for bio-sensing applications.
c. Environmental protection
One of the major problems associated with many microstrip microwave resonator-based sensors, is their high susceptibility to the environmental conditions. Since any change in environmental materials could result in variation in the overall effective permittivity (according to Equation (1)), an error occurs in the resonance frequency. For overcoming this issue, a simple and yet very efficient method is presented in this work. Specifically, a thin metallic plate is placed in parallel with the sensor tag and acts as an isolation layer that reduces the impact of external materials and their variations to the sensor. Fig.4 presents the proposed plate alongside the overall field distribution of the sensor as well as its transmission response. It could be seen that the field is drastically decayed outside of the isolation layer, indicating an insensitivity of the proposed structure to the materials that are located outside of the isolation layer.
To investigate further the impact of the added isolation layer, another simulation is performed that reports the frequency shift as the result of proximity of a material with relative permittivity of εr=35 and the overall conductivity of 1S/m. As illustrated in Fig.5(a), the result of this simulation indicates the importance and the influence of the isolation layer on the reliability and robustness of the structure. It is interesting to notice that the mentioned permittivity in the simulation of Fig.5 could be happened simply by touching the earlobe with a fingertip. Since the structure is protected by the ground plane on the other side, the presented sensor could be identified as an environmentally robust structure.
Another interesting aspect of the isolation layer is the impact of the distance between the isolation layer and the tag. As the distance between the isolation layer and the tag decreased, the quality factor of the sensor is reduced due to more energy absorption by the iso-potential plane created from the isolation layer. Converesely, larger distances between the tag and the isolation layer reduce the user’s comfort and could even make the device impractical. Therefore, there is a trade-off between the user’s comfort and the quality factor of the sensor. This concern is studied in Fig.6 showing the response of the sensor at various tag-isolation layer distances. At reasonable distances, acceptable quality factors are achievable.
a. Dehydration modelling
Maintaining the in-vitro experiments requires the microwave modeling of the TUT which is earlobe in here. For that extent, dehydration, and its impacts on the fluids compartments in the TUT is also needed to be considered. Although the skin is affected by dehydration, specifically the dermis layer which contains ~70% of water [41], shaved mouse skin is used for the model where the water content of the skin is considered for the calculation of dehydration effects as well.
Table II. Concentration of the electrolytes utilized in all the experiments.
|
Material |
Concentration mM/lit |
Material |
Concentration mM/lit |
|
Na+ |
140 |
Ca2+ |
1.5 |
|
Cl- |
105 |
Mg2+ |
1 |
|
HCO3- |
25 |
HEPES (buffer) |
5 |
|
K+ |
5 |
HPO42+ |
1.2 |
As dehydration is calculated as the percentage of total body water loss [42], the overall percentage of water within the model system is also changed. In addition, as ~7% of the material inside the earlobe is blood, the model system contains the same percentage of horse plasma. As water loss in plasma is about 20% of the dehydration rate [43], the same amount is applied here as well. As shown in Table I, about 45% of the earlobe is an adipose tissue consisting of about 70% fat and 30% water. According to these percentages and the overall thickness of the model, a 1mm fat layer from a mouse is also included in the model alongside the skin wrapped inside a small plastic bag. The composition of the interstitial fluid (ISF) listed in Table II and are considered as the electrolytes within the water content of the earlobe. It is known that the water loss due to dehydration occurs more in ISF than intracellular fluid [44]. Since ISF consists about 26% of the total body water, the same percentage in the electrolyte variations is applied to the fluid inside the model. According to [45], for 2.2% dehydration in rats, about 0.9% of Na+, and 1.1% of Cl-, as the main ions in ISF, is lost meaning increasing the electrolyte concentration in the model after dehydration. The same percentage and rate is applied to the model as well.
One of the most challenging problems associated with building this model for dehydration is modelling the volume reduction while keeping the sensors untouched. Any movements of the sensor components during the experiments could result in significant experimental errors. On the other hand, it is very difficult to precisely control the volume and especially the thickness of the model for dehydration. For example, 1% dehydration, results in thickness reduction of about 10-20µm which is not easily manageable in the lab environment. To address this issue, we replaced the reduced water with a material that could be solved in water while having relatively similar electromagnetic characteristics as the tissues in the sample. Due to its good electromagnetic features, ethanol is used as a replacing fluid for dehydration experiments. According to the high difference between dielectric permittivity of water and ethanol, it doesn’t make significant errors in the measurements. In addition, since the conductivity of the liquids inside the model is very high due to relatively high concentration of ions, ethanol’s small conductivity does not play an important role in the measurements.
b. Experimental setup
Fig.7 presents the experimental setup including the earlobe model and the fabricated sensor. All the measurements are conducted using a S5065 VNA from Coppermountain Technologies with the calibration range of 900 MHz-1.2GHz, measuring points number of 8001, and IF bandwidth of 1KHz. The volume of the exposed samples is 400μlit. The thickness of the model including the container and fat and akin layers is 6mm which is close enough to the average actual earlobe thickness minus the thickness of the containers and wraps walls. The utilized horse serum is provided from Thermofisher©. All the water samples are prepared from deionized (DI) double distilled water with 18MΩ.cm of resistance. The container wall thickness is about 1mm and with εr= 4.3 and tanδ=0.0047. All the samples are prepared precisely with a 10µlit. precision dispenser.
To investigate the impact of water replacement with ethanol, Fig. 8 presents the transmission response of the sensor with either the empty container, a container filled with DI water, or the container filled with ethanol. With respect to considering the transmission zero, the response of the sensor to ethanol exhibits less frequency shift and less amplitude suppression as expected. Therefore, selection of ethanol as water replacing material for dehydration tests is reasonable.
c. Results and discussion
To study the stability of the response of the sensor, the transmission zero is measured over the time. As can be seen from Fig.9, the sensor presents an stable response with variation in the range of ±40KHz. This error is less than 0.5% of recorded dehydration variation and therefore represents a minor error in the performance of the sensor. To test the robustness and repeatability of the measurements, a return to zero experiment was performed by alternating between 0 and 20% dehydration. As illustrated in Fig.10, the sensing setup presents a very high repeatability as all the samples result in the same response. Fig.11 presents the frequency shift versus dehydration percentages. It could be seen that, due to the highly sensitive design of the proposed dehydration system, as high sensitivity as 95KHz/1% of dehydration is achieved.
To obtain a practical insight into the robustness of the sensor against the environmental variations, another experiment is performed, whereby a material with relative permittivity of about 35 is moved to a closer proximity of the isolating copper layer located at the overall distance of 3mm to the sensing tag. The container is filled with the experimental fluid related to 0% dehydration. The same approach with the same distance to the tag is accomplished without the isolating copper layer and as presented in Fig.12, a significant error is prevented by addition of the isolation layer. This error, according to the information presented in Fig.11, is more than the extreme case of dehydration level (20%) studied in this research. These results emphasize the importance and benefits of the addition of the isolating copper layer.
A novel chipless passive microwave resonator-based dehydration sensor has been proposed. The sensor is working based on electromagnetic coupling between a gap coupled transmission line as the reader and a chipless split ring resonator as the tag sensor. The organ under the test, here earlobe, has been sandwiched between the reader and the tag where the due to the high field concentration attain as high sensitivity as 100 (KHz/1% of dehydration level). The experimental results verify the effectiveness of the proposed sensor for the measurement of dehydration. The sensor is capable of reliably monitoring dehydration levels quantitatively in a non-invasive way. Our research on the development of non-invasive microwave sensors for biomedical applications is ongoing.
Data availability
The sensor model data and the measurement results are available from the corresponding author(s) on reasonable request. Data access is restricted to research and educational applications.
- D. R. Thomas et al., “Understanding clinical dehydration and its treatment,” J Am Med Dir Assoc, vol. 9, no. 5, pp. 292–301, 2008.
- S. G. Albert, B. R. S. Nakra, G. T. Grossberg, and E. R. Caminal, “Vasopressin response to dehydration in Alzheimer’s disease,” J Am Geriatr Soc, vol. 37, no. 9, pp. 843–847, 1989.
- R. Buffa, R. M. Mereu, P. F. Putzu, G. Floris, and E. Marini, “Bioelectrical impedance vector analysis detects low body cell mass and dehydration in patients with Alzheimer’s disease,” J Nutr Health Aging, vol. 14, no. 10, pp. 823–827, 2010.
- W. L. Kenney and P. Chiu, “Influence of age on thirst and fluid intake,” Med Sci Sports Exerc, vol. 33, no. 9, pp. 1524–1532, 2001.
- L. Hooper et al., “Clinical symptoms, signs and tests for identification of impending and current water-loss dehydration in older people,” Cochrane Database of Systematic Reviews, no. 4, 2015.
- C. R. Gross, R. D. Lindquist, A. C. Woolley, R. Granieri, K. Allard, and B. Webster, “Clinical indicators of dehydration severity in elderly patients,” J Emerg Med, vol. 10, no. 3, pp. 267–274, 1992.
- A. M. El-Sharkawy, O. Sahota, R. J. Maughan, and D. N. Lobo, “The pathophysiology of fluid and electrolyte balance in the older adult surgical patient,” Clinical Nutrition, vol. 33, no. 1, pp. 6–13, 2014.
- L. E. Armstrong, J. A. H. Soto, F. T. Hacker, D. J. Casa, S. A. Kavouras, and C. M. Maresh, “Urinary indices during dehydration, exercise, and rehydration,” Int J Sport Nutr Exerc Metab, vol. 8, no. 4, pp. 345–355, 1998.
- N. Bohnen, D. Terwel, M. Markerink, J. A. Ten Haaf, and J. Jolles, “Pitfalls in the measurement of plasma osmolality pertinent to research in vasopressin and water metabolism,” Clin Chem, vol. 38, no. 11, pp. 2278–2280, 1992.
- J. C. K. Wells, P. S. W. Davies, M. S. Fewtrell, and T. J. Cole, “Body composition reference charts for UK infants and children aged 6 weeks to 5 years based on measurement of total body water by isotope dilution,” Eur J Clin Nutr, vol. 74, no. 1, pp. 141–148, 2020.
- A. M. Silva et al., “Usefulness of raw bioelectrical impedance parameters in tracking fluid shifts in judo athletes,” Eur J Sport Sci, pp. 1–10, 2019.
- F. Trenz, R. Weigel, and A. Hagelauer, “Methods for Human Dehydration Measurement,” Frequenz, vol. 72, no. 3–4, pp. 159–166, 2018.
- D. P. Rose et al., “Adhesive RFID sensor patch for monitoring of sweat electrolytes,” IEEE Trans Biomed Eng, vol. 62, no. 6, pp. 1457–1465, 2014.
- D. Morris, S. Coyle, Y. Wu, K. T. Lau, G. Wallace, and D. Diamond, “Bio-sensing textile based patch with integrated optical detection system for sweat monitoring,” Sens Actuators B Chem, vol. 139, no. 1, pp. 231–236, 2009.
- G. Liu, K. Smith, and T. Kaya, “Implementation of a microfluidic conductivity sensor—A potential sweat electrolyte sensing system for dehydration detection,” in 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, 2014, pp. 1678–1681.
- M. Bariya, H. Y. Y. Nyein, and A. Javey, “Wearable sweat sensors,” Nat Electron, vol. 1, no. 3, pp. 160–171, 2018.
- J. Kim, A. S. Campbell, B. E.-F. de Ávila, and J. Wang, “Wearable biosensors for healthcare monitoring,” Nat Biotechnol, vol. 37, no. 4, pp. 389–406, 2019.
- D. Solovei, P. Businova, J. Drbohlavova, J. Hubalek, V. Adam, and R. Kizek, “Non invasive possibility of body dehydration monitoring,” in SENSORS, 2011 IEEE, IEEE, 2011, pp. 1113–1116.
- C. Liu, F. Tsow, D. Shao, Y. Yang, R. Iriya, and N. Tao, “Skin mechanical properties and hydration measured with mobile phone camera,” IEEE Sens J, vol. 16, no. 4, pp. 924–930, 2015.
- D. C. Garrett and E. C. Fear, “Feasibility Study of Hydration Monitoring Using Microwaves–Part 1: A Model of Microwave Property Changes With Dehydration,” IEEE J Electromagn RF Microw Med Biol, vol. 3, no. 4, pp. 292–299, 2019.
- D. C. Garrett, J. R. Fletcher, D. B. Hogan, T. S. Fung, and E. C. Fear, “Feasibility Study of Hydration Monitoring Using Microwaves–Part 2: Measurements of Athletes,” IEEE J Electromagn RF Microw Med Biol, vol. 3, no. 4, pp. 300–307, 2019.
- A. Mason et al., “Noninvasive in-situ measurement of blood lactate using microwave sensors,” IEEE Trans Biomed Eng, vol. 65, no. 3, pp. 698–705, 2017.
- M. Benlamri et al., “Planar microwave resonator with electrodeposited ZnO thin film for ultraviolet detection,” Semicond Sci Technol, vol. 35, no. 2, p. 25003, 2019.
- N. Hosseini, M. Baghelani, and M. Daneshmand, “Discrete Microwave Spectroscopy using Planar Resonator,” in 2019 IEEE Canadian Conference of Electrical and Computer Engineering (CCECE), IEEE, 2019, pp. 1–4.
- V. Rawat, S. Dhobale, and S. N. Kale, “Ultra-fast selective sensing of ethanol and petrol using microwave-range metamaterial complementary split-ring resonators,” J Appl Phys, vol. 116, no. 16, p. 164106, 2014.
- E. L. Chuma, Y. Iano, G. Fontgalland, and L. L. B. Roger, “Microwave sensor for liquid dielectric characterization based on metamaterial complementary split ring resonator,” IEEE Sens J, vol. 18, no. 24, pp. 9978–9983, 2018.
- M. Baghelani, Z. Abbasi, and M. Daneshmand, “Noncontact high sensitivity chipless tag microwave resonator for bitumen concentration measurement at high temperatures,” Fuel, vol. 265, p. 116916, 2020.
- Z. Abbasi, M. Baghelani, M. Nosrati, A. Sanati-Nezhad, and M. Daneshmand, “Real-Time Non-Contact Integrated Chipless RF Sensor for Disposable Microfluidic Applications,” IEEE J Electromagn RF Microw Med Biol, 2019.
- M. Nosrati, Z. Abbasi, M. Baghelani, S. Bhadra, and M. Daneshmand, “Locally Strong-Coupled Microwave Resonator Using PEMC Boundary for Distant Sensing Applications,” IEEE Trans Microw Theory Tech, vol. 67, no. 10, pp. 4130–4139, 2019.
- M. Baghelani, Z. Abbasi, M. Daneshmand, and P. E. Light, “Non-invasive Lactate Monitoring System Using Wearable Chipless Microwave Sensors with Enhanced Sensitivity and Zero Power Consumption,” IEEE Trans Biomed Eng, vol. 69, no. 10, pp. 3175–3182, 2022.
- M. Baghelani, Z. Abbasi, M. Daneshmand, and P. E. Light, “Non-invasive continuous-time glucose monitoring system using a chipless printable sensor based on split ring microwave resonators,” Sci Rep, vol. 10, no. 1, pp. 1–15, 2020.
- M. R. Islam et al., “Metamaterial sensor based on reflected mirror rectangular split ring resonator for the application of microwave sensing,” Measurement, vol. 198, p. 111416, 2022.
- A. R. Cerny et al., “Development of Flexible Moisture Sensors Based on the Corrosion and Degradation of Conductive Substrates,” ACS Appl Electron Mater, vol. 4, no. 9, pp. 4340–4356, 2022.
- Y. Cao, C. Ruan, K. Chen, and X. Zhang, “Research on a high-sensitivity asymmetric metamaterial structure and its application as microwave sensor,” Sci Rep, vol. 12, no. 1, p. 1255, 2022.
- A. N. Patel, T.-P. Jung, and T. J. Sejnowski, “A wearable multi-modal bio-sensing system towards real-world applications,” IEEE Trans Biomed Eng, vol. 66, no. 4, pp. 1137–1147, 2018.
- A. Waugh and A. Grant, Ross & Wilson Anatomy and physiology in health and illness E-book. Elsevier Health Sciences, 2014.
- C. Gabriel, S. Gabriel, and y E. Corthout, “The dielectric properties of biological tissues: I. Literature survey,” Phys Med Biol, vol. 41, no. 11, p. 2231, 1996.
- S. Gabriel, R. W. Lau, and C. Gabriel, “The dielectric properties of biological tissues: II. Measurements in the frequency range 10 Hz to 20 GHz,” Phys Med Biol, vol. 41, no. 11, p. 2251, 1996.
- S. Gabriel, R. W. Lau, and C. Gabriel, “The dielectric properties of biological tissues: III. Parametric models for the dielectric spectrum of tissues,” Phys Med Biol, vol. 41, no. 11, p. 2271, 1996.
- E. Porter, J. Fakhoury, R. Oprisor, M. Coates, and M. Popović, “Improved tissue phantoms for experimental validation of microwave breast cancer detection,” in Proceedings of the fourth European conference on antennas and propagation, IEEE, 2010, pp. 1–5.
- M. GUZMÁN-ALONSO and T. M. CORTAZÁR, “Water content at different skin depths,” Discover GlucoPure™, p. 35, 2016.
- F. Ashcroft, Life at the extremes: the science of survival. Univ of California Press, 2002.
- L. C. Senay Jr, “Temperature regulation and hypohydration: a singular view,” J Appl Physiol, vol. 47, no. 1, pp. 1–7, 1979.
- M. J. Durkot, O. Martinez, D. Brooks-McQuade, and R. Francesconi, “Simultaneous determination of fluid shifts during thermal stress in a small-animal model,” J Appl Physiol, vol. 61, no. 3, pp. 1031–1034, 1986.
- H. Nose, G. W. Mack, X. Shi, and E. R. Nadel, “Shift in body fluid compartments after dehydration in humans,” REPLACEMENT AND HEAT STRESS, 1994.
No competing interests reported.