Grigore

Popa University o{Medicine and Pharmacy, ia:ji, Romania, November 21-23,2013

The 4th IEEE International Conference on E-Health and Bioengineering - EHB 2013
T

Wireless Skin Temperature Measurement System for

Circadian Rhythm Monitoring
Cristian Rotariu, Hariton Costin, Alexandru Pasarica,

Bogdan Dionisie

Ciprian Cristea

Department of Physiology
Grigore T. Popa University of Medicine and Pharmacy
Iasi, Romania

Department of Biomedical Sciences

Grigore T. Popa University of Medicine and Pharmacy
[asi, Romania
cristian.rotariu@bioinginerie.ro

Ab stract-Nowadays a significant part of the population

suffers of sleep disorders, so the dynamic long term monitoring
of skin temperature during sleep is very useful for analysis and
understanding the circadian rhythm. In this paper we describe
the realization of a patient skin temperature monitoring system,
based on wireless devices, capable to measure and transmit the
patient's skin temperature to a central monitoring station. The
use of the proposed system is suitable for remote continuous
long-time patient monitoring, in-hospital or at-home, as a part of
a

diagnostic

system

uses

measurement

procedure

or

custom
devices

during

recovery.

developed
that

wireless

perform

the

The

described

temperature
temperature

measurements and wirelessly transmit them to the monitoring

station.

A graphical user interface running on the central

monitoring

station

was

developed,

measurements and alerts,

used

for

displaying

the

when patient temperature values

exceed the preset limits.

Keywords-circadian

rhythm,

temperature, wireless sensors.

I.

patient

monitoring,

skin

INTRODUCTION

Re[iab[e long term patient monitoring is crucial for a

number of medical conditions requiring circadian rhythm
analysis [ 1], sleep-related disorders, hypertension, ischemic
heart diseases, heart failures, and strokes. It is well known
that sleep and body temperature are closely related and the
human performances depend on body temperature [2].
Nowadays more and more people suffering from circadian
rhythm sleep disorders have their timing of sleep affected,
and usually are unable to sleep and wake at the times required
for daily normal activities.
Wireless sensor networks (WSN) have great potential for
continuous monitoring at hospital or even patient's home,
early detection of abnormal conditions, and/or supervised
rehabilitation [3]. The advances technologies are driven by
the developments in wireless communications and wearable
computing completed by the minimization of weight and size
of sensors, portability, connectivity, reliability and easy
integration into complex systems. The availability of low
power short range RF transceivers in the unlicensed band
allows the design of a small, low-cost, battery operated
wireless sensors [4].

On the market, there is a significant increase in the number

of various wireless temperature acquisition devices. Although
these devices are used only to collect data, they still remains
the most used. Data processing and analysis are performed
offline, making them impractical for continual long term
monitoring.
Monitoring the patient's temperature in hospital or at-home
requires the use of temperature sensors attached by wires to
medical devices, which limits the patient's movements. For
example, a commonly used method uses for temperature
measurement thermistors attached by wires to the temperature
data logger. Although acceptable for shorter periods of time,
this method is unacceptable for long term monitoring or
monitoring of sleeping patients. [n order to avoid these
situations, we use WSNs, based on low power devices,
having RF transceivers.
The patient's skin temperature is continuously measured by
the proposed system and it is wirelessly transmitted to the
central monitoring station using a standard low power WSN,
avoiding using more expensive Bluetooth or WiFi nodes.
This paper describes a temperature remote monitoring
system for circadian rhythm based on wireless sensor nodes.
The sensor nodes contain commercially available temperature
sensors and wireless modules, perform the temperature
measurements and wirelessly transmit them to the monitoring
station. The central monitoring station runs a patient
temperature monitor application that displays the patient
temperature and alerts, when patient temperature values
exceed the preset limits.
[I.

MATERIALS AND METHODS

A conceptual view of the proposed system, as represented

in Fig. 1, contains the following main components: a WSN of
temperature acquisition devices, one sensor node for each
monitored patient attached on his/her chest, an Access Point
as a network coordinator, and a PC as a monitoring station.
Each temperature acquisition device contains an
eZ430RF2500 wireless module (Fig. 2) [5]. The eZ430RF2500 is a complete wireless development module for the
MSP430F2274 microcontroller [6] and CC2500 wireless RF
transceiver [7] that includes the HW and SW resources
required to develop small WSN applications.

978-1-4799-2373-1/13/$31.00 2013 IEEE

Access point

Patient
-

Central monitoring station

[), .
%l l
Sn.or
.. od.

Ir-="'-JI '\

Fig. I. Wireless Skin Temperature Measurement System

The MSP430F2274 is a l6-bit RISC ultra-low-power mixed

signal microcontroller, optimized to achieve extended battery
life in portable measurement applications. [t has an internal
memory structure of 32 kb flash memory and 1kb on-chip
static RAM. Other features of the MSP430F2274 include:
two internal 16-bit timers, a fast [0 bit 200 kbps AID
converters with internal reference, enhanced universal serial
communication interfaces USART, I2C, SPI. The internal
digitally controlled oscillator allows wake-up from low
power modes to active mode very fast, facility that is used to
extend batteries life. Typical applications of MSP430F2274
include sensor systems that capture analog signals, convert
them to digital values, and then process the data for display or
for transmission to a host system. Stand-alone radio
frequency sensor front ends are another area of application.

rate up to 500 kBaud, and consumes less than 2 1.2 mA in

transmission mode at 0 dBm output power and [7.0 rnA in
receiving mode. The CC2500 is connected to the
MSP430F2274 microcontroller by using the SP[ interface and
has an indoor line-of-sight range up to 50 meters.
We used the SimpliciTi protocol (Fig. 3) from Texas
Instruments to transfer data from sensor node to central
monitoring station [8]. SimpliciTI is an open source low
power network protocol used with battery-operated devices.
As wireless network protocol, SimpliciTI supports End
Devices in a peer-to-peer network topology, an Access Point
as a network coordinator to store and forward the measured
temperature to the PC, and Range Extenders used to extend
the range of the wireless network.

Access point

Range Extender

End Device

Sensor
Node

Sensor
Node

Sensor
Node

Sensor
Node

Fig. 2. The eZ430-RF2500 wireless module

Fig. 3. The SimpliciTi wireless protocol

The CC2500 is a low-cost 2.4 GHz wireless RF transceiver

designed for low-power wireless applications. The circuit
uses for data transmission the ISM (Industrial, Scientific and
Medical) and SRD (Short Range Device) frequency bands,
supports various modulation formats, has a configurable data

For the skin temperature measurement we use TMP275

temperature sensor (Texas Instruments). The TMP 275 is a
high precision 0.5 C accurate, two-wire temperature sensor
with serial output, capable of reading temperatures with a
resolution of 0.0625C. The TMP275 is directly connected to

the eZ430-RF2500 using the standard I2C communication

interface (Fig. 4) and requires no external components for
normal operation except for resistors on SCL and SDA lines.

VCC
Rl

Rl .
10K

vee

SCL

VCC

Ul

10K

SDA

Rend battery voltage

1l'C2 lOOnFCl
I

l !ThRT v I

uF

TMP275

GNo

transceiver was minimized by sending it into a low power

sleep mode when it is not in use.

P':fSP430F2274

Fonnat smnples
for trllslllissioll

DV.s

Selld dt to

Fig. 4. The electrical interface between TMP275 and MSP430F2274

The accuracy of the TMP275 temperature sensor for the

35--45C temperature interval is below O.2C (Fig. 5) and
the conversion of 12 data bits took place in 220ms.
TEMPERATURE ERROR

vs

TEMPERATURE

0.500
0.375

0250

0.12 5

Fig. 7. The flowchart of firmware working on MSP430F2274

Q; -0.125

-0 . 250

I---

-0.375

A user application interface (Fig. 8) running on the central

monitoring station, was developed to run on Windows based
operating
systems,
and
was
written
using
the
LABWindows/CVI programming environment [9].

-0.500
-55

-35

-15

25

45

Temperature

65

( " C)

85

105

125 130

Fig. 5. The TMP275 temperature error vs temperature

Temperature MONITOI\

Patient Marne

IMihailescu Ion

Eower

00

The prototype of Sensor Node is presented in Fig. 6.

Power Meter

TEMP

22.10.201313:18 Mihailescu Ion FEVER!

22.10.201313:1 9 Mihailescu Ion FEVER!
22.10.201313:20Mihailescu Ion FEVER!
22.10201313:21 Mihailescu Ion FEVER!
1l.10201313:22Mihaile3CU Ion FEVER!

I

I

,..JOn

Con/ioLre

...
S.al/e

'"
>I""
(!ol

Fig. 6. The temperature acquisition device - Sensor node

In Fig. 7 is represented the flowchart of firmware working

on MSP430F2274 microcontroller from the eZ430-RF2500
module. In this instance, the MSP430F2274, after a START
command, wakes up to measure the patient skin temperature,
sensor battery voltage, and communicate the result to central
monitoring station. Energy consumption by the wireless RF

Fig. 8. The Graphical User Interface of application running on the central

monitoring station

The main section of interface consists of a drawing area for

displaying the patient's temperature, a control area consisting
of various buttons and indicators (sensor battery voltage), and
a displaying area for user-specific information such alerts.
The control area consists of buttons like Open, Save, Start
and Stop temperature recording, Quit application, and a
Power switch (ON/OFF).
The "Configure" button activates the Configuration Panel
and allows user to select the RS232 communications
parameters between the eZ430-RF2500 receiver module and
central monitoring station with several options: COM Port,
Baud Rate, Parity, Data Bits and Stop Bits. The default values
for these parameters are presented in the Fig. 9.
G:J

COM
COM1COM2COM3COM4COM5COM6COM7COM&-I

under the curve of

1450flV*s (Fig. 1 1). In this way, we
theoretically computed an average current:
average_current
period_oCtransmission

110120019200-

300600-

2400 4300-

384009600 57600

evenmark n =space

el

6-5-1
&- -

Data Bos

Parity

iFi)

1-

7-

hours_oCoperation
currentJating / average_current
240 [mA*hr] / 0.0 145 [mAl
16552 [hrs] / 24 [hrs/day]
690 [days].
=

RIGOL
STOP

Stop 80s
2-

(measured_voltage
/
10Q)
1450flV*S /10 Q /10 s 14.5 flA.

In order to compute the battery life expectancy, assuming

that the battery still maintain a 240mA*hr until its capacity is
exhausted, we obtain:

Co n figure RS232 Port

Port

Oose

FiK ge The Configuration Panel

L!GIl!!!I50.0mU

Time

1.000ms

0>200.0us

Fige Se The current profile of sensor node

III.

RESULTS

A prototype system containing all described above has been

implemented and tested. Using wireless sensor nodes attached
on patient's chest we have successfully making patient's
temperature measurements and forward the measured values
to the central monitoring station through WSN. The sample
time used to acquire the patient temperature was chosen at 10
sec.
In order to analyze the current profile of the wireless sensor
node, the hardware used is represented in the Fig. 10.
Rigol DS5022M

Oscilloscope

Power Supply

HAMEG

Access Point
eZ430-RF2500

HMS040-3
+ 3V

J:--=Vr'"

IV.

This paper describes a prototype of a remote patient skin

temperature monitoring system used for circardian rhythm
monitoring, based on wireless sensor networks, capable to
measure and transmit the patient's skin temperature to a
central monitoring station.
The use of described system is suitable for continuous long
time multiple patient monitoring, in-hospital or at-home.
The system uses custom developed wireless temperature
measurement devices and is implemented using commercially
available temperature sensors and wireless modules.
REFERENCES
[I]

[2]

100 Resistor

r'-

Sensor Node
eZ430-RF2500

'V

-f-

Fige lOe Sensor node power supply test hardware

The largest contributor to current consumption IS the

CC2500 RF transceiver. To calculate the average current
consumption by the sensor node, we transferred the data
acquired by Rigol DS5022M digital oscilloscope to a
computer and, with the aid of MATLAB software [ 10] we
computed the integral of the voltage curve - resulted an area

CONCLUSIONS

[3]

[4]
[5]
[6]
[7]
[S]
[9]
[10]

Jovanov Ee, Raskovic Dc, Honnigo Re, 'Thermistor-Based Breathing

Sensor for Circadian Rhythm Evaluation", Biomedical Sciences
Instrumentation, VoL 37, ppe 493-497, 2001
Wright KY, lL, Hull JT, Czeisler CA, "Relationship between
alertness, performance, and body temperature in humans" , Am J
Physiol Regul Integr Comp PhysioL Dec;2S3(6):R1370-7, 2002
Jovanov R, "A Survey of Power Efficient Technologies for Wireless
Body Area Networks," Proce of the 30th Annual International
Conference of the IEEE Engineering in Medicine and Biology Society,
Vancouver, Canada, ppe 362S, 200Se
Bozomitu Re Ge, "Radio transmitters and radio receivers",
Performantica Publishing House, Iasi, Romania,2012
eZ430RF2500
Development
Tool
User's
Guide,
http://wwwetiecom/tool/ez430-rf2500
http://wwwetiecomllitlds/symlinklmsp430f2274epdf
http://wwwetiecom/lit/ds/symlink/cc2500epdf
http://wwwjiecom/tool/simpliciti
http://wwweniecom
http://wwwemathworksecom