EMBC2016_Huang

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309340503

Wearable deep body thermometers and their uses in continuous monitoring for
daily healthcare

Conference Paper in Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society.
Conference · August 2016
DOI: 10.1109/EMBC.2016.7590669

CITATIONS READS
9 792

5 authors, including:

Ming Huang Toshiyo Tamura


Nagoya City University Waseda University
91 PUBLICATIONS 743 CITATIONS 405 PUBLICATIONS 7,185 CITATIONS

SEE PROFILE SEE PROFILE

Shigehiko Kanaya
Nara Institute of Science and Technology
492 PUBLICATIONS 18,947 CITATIONS

SEE PROFILE

All content following this page was uploaded by Toshiyo Tamura on 12 November 2018.

The user has requested enhancement of the downloaded file.


Wearable Deep Body Thermometers and Their Uses in Continuous
Monitoring for Daily Healthcare
Ming Huang, Member, IEEE, Toshiyo Tamura*, Senior Member, IEEE, Takumi Yoshimura, Tadahiro
Tsuchikawa and Shigehiko Kanaya

Abstract— This paper introduces noninvasive deep body adhere to strict rules protecting those who work outdoors. In
thermometers suitable for continuous deep body temperature this context, heat strain control is essential, especially in
(DBT) measurement. On the basis of their features, they were summer. In Japan, workers are required to wear a long-sleeved
used in DBT monitoring for daily healthcare. A thermometer jacket and trousers while working. This makes heat dissipation
based on the dual-heat-flux method (T_DHFM), and an aural more difficult. There are thus strong demands for the close
canal thermistor (ACT), were used in two studies of daily
monitoring of thermal regulation. Another example of the use
healthcare. The medical device CoreTemp by Terumo, based on
the zero-heat-flux method, was also used for a DBT reference. of physiological monitoring is that of people with spinal cord
The first study focused on preventing heat stroke in a injuries. For paraplegics and quadriplegics, temperature
high-temperature and high-humidity environment, while the receptors in the body cannot communicate with the
other focused on the temperature monitoring of patients with hypothalamus, which induces poikilothermia. However, in
spinal cord injuries. In the first study, CoreTemp and T_DHFM these individuals, outdoor activities are recommended to
were used, whereas T_DHFM and ACT were used in the second improve their quality of life; therefore, when outdoors, their
study. Using the results from these two studies, we discuss the body temperature should be monitored appropriately.
availability and performance of each thermometer and indicate
the necessity of an appropriate method of measuring DBT.
In the above scenarios, only noninvasive methods are
I. INTRODUCTION applicable. In the present study, we thus focused on the
following noninvasive methods: use of the medical device
The continuous monitoring of vital signs has revealed an
CoreTemp by Terumo, based on the zero-heat-flux method
increasing amount of information about the human body. It
(ZHFM); use of a thermometer based on the dual-heat-flux
provides time series data, with which trends in physiological
method (T_DHFM); and use of an aural canal thermistor
changes can be identified [1, 2]. The shift in measuring vital
(ACT). Among these, CoreTemp and ACT have been
signs from a hospital setting to an ambulatory setting and in
approved by the Pharmaceuticals and Medical Devices
daily life has been propelled by the popularity of wearable
Agency of Japan. These three devices will be briefly
devices. Starting with a wearable pulse oximeter, blood
introduced later.
pressure and deep body temperature (DBT) can now be
measured continuously using lightweight devices.
We carried out studies on the above two scenarios using
Among the vital signs, DBT is inherently difficult to practical experiments. For heat stroke prevention, trials were
measure noninvasively. It refers to the temperature inside the carried out in a high-temperature and high-humidity
natural cavities of the human body and thus requires invasive environment, while the temperature monitoring of
measurement. Rectal temperature is generally used as a gold paraplegics during outdoor activities was carried out in winter.
standard in a hospital setting. However, DBT monitoring is Using the findings from these two studies, the availability and
needed not only in hospital but also for healthcare in daily life. performance of each thermometer and the necessity of an
appropriate method of measuring DBT are discussed.
For example, the physiological monitoring of construction
workers is now becoming increasingly important in order to
II. METHODS
Research supported by the Keihanna Science City Healthcare Project of A. Wearable Deep Body Thermometers
the Ministry of Education, Culture, Sports, Science and Technology, Japan,
and partly by Tateishi Science and Technology Foundation, Japan and Health DHFM
and Labour Sciences Research Grants for Comprehensive Research on
Persons with Disabilities, Japan Agency for Medical Research and DHFM is a relatively new method that calculates the DBT
Development (AMED), 2015-2017. based on the heat flux inside a probe. It was originally
M. Huang and S. Kanaya are with Nara Institute of Science and proposed by Kitamura et al. [3]. Using a double heat path
Technology, Ikoma, Nara 630-0192, Japan ;(e-mail: {alex-mhuang, inside a probe, it is possible to calculate the DBT using
skanaya}@is.naist.jp).
T. Tamura is with Osaka Electro-Communication University, Neyagawa, temperature sensors embedded within the probe. The
Osaka 575-0063, Japan (corresponding author Phone +81-72-824-1131 fax fundamental principle of DHFM is illustrated by Fig. 1. A
+81-72-824-0014 e-mail: tamurat@isc.osakac.ac.jp). substrate material with four embedded temperature sensors
T. Yoshimura is with Tokyo Metropolitan college of Industrial constitutes the core of the probe. The substrate material has
technology, Tokyo Japan (e-mail. t-yoshim@metro-cit.ac.jp )
T. Tsuchikawa is with University of Hyogo, Shinzaike-honcho, Himeji, similar physical properties to skin and, when it is attached to
Hyogo, 670-0092, Japan (e-mail: tad@shse.u-hyogo.ac.jp) skin, the heat flow from the body core due to the difference

978-1-4577-0220-4/16/$31.00 ©2016 IEEE 177


(a)
Fig. 1. Illustration of dual-heat-flux method. The substrate material has similar
heat conductivity to skin and thus the heat flow is considered to occur
vertically from the body into the material
between the DBT and the skin temperature mostly occurs into
the substrate material. Additionally, upon implementing heat
isolation peripheral boundary conditions, the heat flows
longitudinally. Since the two heat paths (T1–T3, T2–T4) are
located transversely close to each other, the thermal resistors
(b)
in the skin layer of the two heat paths are the same and, thus, Fig. 2. Design and image of the dual-heat-flux method-based probe used in
one can calculate the DBT using the four sensors (T1–T4) by the studies. The double-cylinder design shown in (a) is used to generate two
the equation below, where k (= R1/R2) is the ratio of heat heat paths with different heat resistors. The prototype (b) is fabricated
resistors inside the probe of the two heat paths. according to the design shown in (a) with four inlaid temperature sensors

(𝑇1 −𝑇2 )(𝑇1 −𝑇3 ) Co. (Tokyo, Japan) and is shown in Fig. 2. Temperature is
𝑇d = 𝑇1 + . (1)
𝑘(𝑇2 −𝑇4 )−(𝑇1 −𝑇2 ) sampled at a frequency of 1 Hz and then transmitted by
Bluetooth to a data logger and saved. The thermistor is
Regarding the accuracy of the original prototype of this embedded in a soft ear-plug and can measure the temperature
method, its results differ by less than 0.1°C from the reference with an error of less than 0.1°C.
CoreTemp thermometer; however, it requires an additional
ZHFM
urethane sponge cover. This method was improved by Huang
ZHFM was originally proposed in the 1970s by Fox and
et al. based on theoretical simulation and experimental
Solman [8]. It involves the use of a noninvasive deep body
validation [4, 5]. By lacking an external heater, its energy
thermometer with good traceability and accuracy comparable
consumption is greatly improved, making this method readily
to those of standard invasive methods. Its fundamental
implementable in a wearable modality.
mechanism is based on the fact that, by heating up a cutaneous
Based on the results of DHFM, we recently developed a probe to the extent that there is no temperature gradient inside
prototype system and tested its basic performance [6, 7]. For it, it can be considered that no heat will flow into the probe
the two studies performed here, as mentioned above, from the contiguous skin and thus no heat will flow from the
prototypes with a height of 15 mm and a radius of 22 mm were interior of the body onto the skin. The DBT will then be equal
used in the pit of the stomach (Fig. 3). to the temperature of the probe. This method was subsequently
improved by Togawa’s group and Terumo Co. (Tokyo, Japan)
ACT
[9]. The thermometer fabricated by Terumo has been
The aural canal is a suitable location for body temperature
approved as a medical device by the Pharmaceuticals and
measurement. Using a thermistor set inside the probe, the
Medical Devices Agency of Japan and its measurements have
temperature of the external aural canal can be determined. If
been shown to correlate well with blood temperature during
appropriate shielding from ambient convection can be
cardiopulmonary operations [10]. The convenience provided
achieved, the temperature measured by this device can be
by this noninvasive thermometer has led to other
regarded as a good approximation of DBT. Additionally,
manufacturers, such as Philips [11,12] and 3M [13], also
concern over perforation of this kind of thermometer is
expending great effort on developing their own sensors based
mitigated by the coating of the thermistor with soft material.
on ZHFM.
The ACT that we used is manufactured by Nikkiso-Therm
Thanks to the adoption of the external heater embedded
inside the probe, ZHFM shows a rapid response to the change
of DBT with stable readings; however, the power
consumption has also restricted this method from being used
in daily life. Using this kind of thermometer, it is only possible
to perform the continuous monitoring of DBT in a hospital
setting.
B. Experimental Studies
Fig. 2 The aural canal thermistor used in the study of temperature monitoring of
a patient with spinal cord injury. The aural canal thermistor that we used is The performances and availabilities of these two wearable
manufactured by Nikkiso-Therm Co. thermometers were applied to the studies in environments

178
experienced in daily life. The first study focused on the outdoor temperature of 10.2°C. A patient (with a spinal cord
prevention of heat stroke in a high-temperature and injury) and a healthy control underwent the same protocol. In
high-humidity environment. In this experiment, both the first phase, both subjects stayed in a room with the
CoreTemp and T_DHFM were applied. In addition, the temperature at 20°C for 4 h (14:00–18:00), after which they
surface temperatures (ST) inside and outside a jacket were both went out to a park for about 2 h and then returned to the
monitored and regarded as potential parameters for judging room at 20:00. The experiments were approved by the Ethics
the thermal state of the human body. Committees of Nara Institute of Science and Technology,
Osaka Electro-Communication University, and the University
Six young male subjects (age: 25.3±8.9 years, height:
of Hyogo. The informed consents were received from all
169.6±5.2 cm, weight: 64.7±8.6 kg) participated in these
subjects.
experiments. They underwent experiments in an isothermal
chamber under conditions of ambient temperature of 40°C and III. RESULTS
relative humidity of 40%. Each experiment comprised three
Our studies were aimed at examining the performance of
phases: acclimation (10 min), cycling exercise (20 min, 50 W),
the three methods for practical applications in terms of their
and recovery (10 min).
availability and stability. For the study on preventing heat
DBT at the forehead and back and ST inside and outside a stroke, we evaluated the availability of T_DHFM in an
jacket were measured using a noninvasive thermometer environment in which the temperature is higher than the DBT.
(CoreTemp CM-210; Terumo). CM-210 recorded all of the Since CoreTemp has a maximum working temperature of
temperature data at intervals of 2 s. In addition, T_DHFM was 40°C with 0.1°C error, it was regarded as a reference against
applied at the pit of the stomach, with sampling at intervals of which to evaluate T_DHFM. The justification for using the
surface temperature as the only criterion representing the
thermal state was also examined.
The measurements of ST and DBT by CoreTemp are
summarized in Table 1, for which the mean values are shown.
For each kind of measurement, two-sample t-test was used to
test the relationship between two situations, namely, the
cycling phase and the recovery phase, whereas the acclimation
phase data were not analyzed. From Table 1, it can be seen that
there was no significant difference for ST, while the DBTs at
the forehead and back were both significantly higher than
during the cycling phase.

TABLE I. SUMMARY OF THE TEMPERATURE MEASUREMENT IN THE


Fig. 4. The set-up of CoreTemp and the dual-heat-flux method. The probes STUDY ON PREVENTING HEAT STROKE
of CoreTemp were fixed with surgical tape, while the device for the
dual-heat-flux method was fixed to the pit of the stomach using a supporter ST (Mean ± SD °C) CoreTemp (Mean ± SD °C)
belt.. Situ.
Inside Outside Back* Forehead*
1 s. The set up of the probes is shown in Fig. 4.
DE 38.36 ± 0.19 38.96 ± 0.23 37.04 ± 0.18 37.24 ±0.09
The second study focused on monitoring the temperature
of patients with spinal cord injuries, in which T_DHFM and AE 38.39 ± 0.12 38.94 ± 0.13 37.29 ± 0.03 37.38 ±0.03
ACT were used. CoreTemp is not suitable here since it is Situ. is the condition of the subject; DE denotes the situation during exercise by cycling; AE denotes
the situation after the exercise; Inside and Outside denote the surface temperatures inside and outside
powered by an alternating current. This experiment was the jacket; SD means the standard deviation and Back and Forehead denote the deep body temperature
carried out based on the conclusion that, by T_DHFM, it is measurements at the back and forehead. An asterisk denotes that there is a significant difference
between the two situations that there is significant difference between the two situation.
possible to measure DBT change in the thorax and, thus, to The availability of T_DHFM was examined using
compare these two methods experimentally. The experiment CoreTemp as a reference. Since T_DHFM is a passive method,
lasted 6 h, from 14:00 to 20:00 in late fall, with an average it took longer (by ~20 min) to achieve heat equilibrium than
Temperature °C

Time (h:min:s)
Fig. 5. Time series recordings by the dual-heat-flux method and the aural canal thermistor in a healthy control (Cont.) and a subject with a spinal cord injury
(Sub.). The protocol consisted of two phases: indoor monitoring (~4 h) and outdoor monitoring (~2 h)

179
with CoreTemp. The average difference between the forehead have been due to the design mitigating the influence of the
(with CoreTemp) and the pit of the stomach (with T_DHFM) environment.
was 0.07°C, while that between the back (with CoreTemp)
and the pit of the stomach (with CoreTemp) was 0.14°C. V. CONCLUSION
By analyzing noninvasive methods for measuring DBT, we
Regarding the monitoring of patients with spinal cord
attempted to meet the practical need for continuous
injury, the availabilities of both thermometers were examined
monitoring of DBT using the CoreTemp medical device,
using time series recordings (Fig. 5). As indicated in the first
T_DHFM, and ACT. The CoreTemp device provided stable
study, T_DHFM required a certain period of time to achieve
measurements but is not really practical in environments
heat equilibrium, but this period was much longer for the
encountered in daily life, while ACT is vulnerable to changes
subject with spinal cord injury (>1 h) than for the control (~30
in the environment, such as temperature changes or the wind.
min).
On the other hand, T_DHFM is capable of providing good
accuracy and is more resistant to environmental influences,
For both thermometers, the outdoor environment potentially making it an ideal substitute for continuous DBT
influenced the measurements, as shown by the data at around monitoring for healthcare in daily life.
18:00. The measured values decreased abruptly when the
subjects went outside, with the exception of the T_DHFM of
ACKNOWLEDGMENT
the target subject. This may have been due to special
measures against getting cold having been taken, which We thank Mr. Uchida for help with the cooling device in
limited the influence of the environment on T_DHFM. the heat stroke experiment..

The influence of the outdoor environment was maintained REFERENCES


throughout the outdoor activities for ACT, which can be seen [1] Y. Chen and W. Chen, “Long-term tracking of a patient’s health
from the recordings of both the control and the target subjects. condition based on pulse rate dynamics during sleep.,” Ann. Biomed.
Eng., vol. 39, no. 12, pp. 2922–34, Dec. 2011.
The recordings of ACT decreased to a level (<36°C) that
[2] D. Weinert and J. Waterhouse, “The circadian rhythm of core
would seldom occur in a thermoneutral person. In contrast, temperature: effects of physical activity and aging.,” Physiol. Behav.,
T_DHFM was rather stable for both of them. vol. 90, no. 2–3, pp. 246–56, Feb. 2007.
[3] K.-I. Kitamura, X. Zhu, W. Chen, and T. Nemoto, “Development of a
new method for the noninvasive measurement of deep body
IV. DISCUSSION temperature without a heater,” Med. Eng. Phys., vol. 32, no. 1, pp. 1–6,
DBT is considered to be a crucial index for representing the Jan. 2010.
thermal status of individuals. However, owing to restrictions [4] M. Huang, T. Tamura, Z. Tang, W. Chen, and S. Kanaya, “Structural
optimization of a wearable deep body thermometer: From theoretical
in the methods for measuring it, this parameter is generally simulation to experimental verification,” Journal of Sensors, 2015, Dec.
only available in a hospital setting. Even for CoreTemp, 2016.
which has shown satisfactory accuracy when compared with [5] M. Huang, T. Tamura, W. Chen, N. Ono, T. Sato, and S. Kanaya,
standard invasive methods [9], it is still difficult to extend its “Evaluation of a Noninvasive Deep Body Thermometer in
Measurement of Specific Positions,” Conf. Proc. IEEE Eng. Med. Biol.
range of applications. Soc., pp. 2395-98, 2015.
Therefore, for the prevention of heat stroke, ST is widely [6] M. Huang, T. Tamura, W. Chen, and S. Kanaya, “Evaluation of
used. However, from the first study in the current work, we structural and thermophysical effects on the measurement accuracy of
determined that the use of this variable alone might be deep body thermometers based on dual-heat-flux method,” J. Therm.
insufficient. When measuring DBT, significant changes were Biol., vol. 47, pp. 26–31, Jan. 2015.
[7] M. Huang, T Tamura, Z Tang, W Chen, S Kanaya, “A Wearable
seen; however, the ST neither inside nor outside the jacket thermometry for core body temperature measurement”, IEEE Journal
reflected these changes. A more reliable index of the internal of Biomedical and health informatics, DOI
thermal state is thus needed and the passive DHFM method 10.1109/JBHI.2016.2532933
may be an ideal substitute for CoreTemp, according to the [8] R. Fox and A. Solman, “A new technique for monitoring the deep body
results of this study. Apart from the good accuracy it showed temperature in man from the intact skin surface,” The Journal of
physiology, vol. 212, no. 2, pp. 8-10, Jan. 1971.
compared with CoreTemp, its low power consumption makes [9] T. Nemoto and T. Togawa, “Improved Probe for a Deep Body
it suitable for a wearable modality. Thermometer,” Med. Biol. Eng. Comput., vol. 26, pp. 456–59, 1988.
[10] M. Yamakage and A. Namiki, “Deep temperature monitoring using a
Temperature measurement in the aural canal is considered zero-heat-flow method,” J. Anesth., vol. 17, no. 2, pp. 108–115, 2003.
a suitable substitute in continuous DBT monitoring. This is [11] L. P. J. Teunissen, J. Klewer, A. de Haan, J. J. de Koning, and H. a M.
Daanen, “Non-invasive continuous core temperature measurement by
especially true for paraplegics, whose skin temperature in the zero heat flux.,” Physiol. Meas., vol. 32, no. 5, pp. 559–70, 2011.
lower limbs differs from that in healthy people. However, our [12] L. Atallah, E. Bongers, B. Lamichhane, and S. Bambang Oetomo,
second study showed that ACT is easily influenced by “Unobtrusive monitoring of neonatal brain temperature using a
zero-heat-flux sensor matrix,” IEEE J. Biomed. Heal. Informatics, vol.
changes in the environment. Using this approach, caregivers 20, no. 1, pp. 100–107, 2016.
would thus have difficulty determining the actual thermal [13] Y. Eshraghi, V Nasr, I Parra-Sanchez, A van Duren, M. Bothaim, T
state of the user. On the other hand, the measurements by Santoscopy, and DI Sessler: An evaluation of a zero-heat flux
T_DHFM were stable, which may have been due to clothes cutaneous thermometer in cardiac surgical patients. Anesth Analg.
119(3):543-9 2014;
acting as a shield against the ambient airflow. It may also
180

View publication stats

You might also like