Wireless Temporal Artery Bandage Thermometer

Wireless Temporal Artery Bandage Thermometer Ivars G. Finvers, James W. Haslett and Graham Jullien Department of Electrical and Computer Engineering University of Calgary Calgary, Alberta, Canada T2N 1N4 Email: Abstract—A bandage based thermometer placed on the tem- ple region of the forehead provides an non-invasive means of measuring a patient’s core body temperature. An array of temperature sensors spaced along the length of the bandage
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Wireless Temporal Artery Bandage Thermometer Ivars G. Finvers, James W. Haslett and Graham Jullien Department of Electrical and Computer EngineeringUniversity of CalgaryCalgary, Alberta, Canada T2N 1N4Email:  Abstract —A bandage based thermometer placed on the tem-ple region of the forehead provides an non-invasive means of measuring a patient’s core body temperature. An array of temperature sensors spaced along the length of the bandage isused to determine the skin temperature over the temporal artery.Temperature sensors buried within the bandage allow the directmeasurement of the heat flow leaving the skin over the temporalartery. By using the method of heat balance in conjunctionwith the skin temperature and heat flux measurements, thetemperature of the blood within the temporal artery, whichis at the core body temperature, can be estimated. Comparedto the traditional hand-held Temporal Artery Thermometer,this approach decreases the sensitivity of the core temperatureestimation to errors arising from air flow across the forehead,perspiration, and the local environment. A wireless link transmitsthe data to a remote monitoring station, allowing long termmonitoring of patient temperature. I. I NTRODUCTION Wireless patient vital sign monitoring systems are under de-velopment with the goal of improving patient health and safety,while minimizing health care costs by reducing medical staff workload. Body temperature was perhaps the first vital signused by physicians to evaluate a patient’s health. Despite this,the accurate non-invasive measurement of body temperature ina continuous fashion remains a difficult problem. Traditionalmethods such as oral, rectal, or tympanic (ear) thermometersare not easily adapted to long term monitoring, while othertechniques such as axillary (armpit) thermometers suffer frompoor accuracy.A relatively new method of non-invasively measuring bodytemperature is the Temporal Artery Thermometer (TAT) whichwas invented by Pompei in 1998 [1]–[3]. A hand-held infraredtemperature probe, similar in physical construction to an earthermometer, is scanned across the temporal artery on theforehead and core body temperature is estimated from the skintemperature measured directly over the artery.The skin acts as a radiator element within the body’sthermoregulation system. The blood flow (perfusion) to thecapillary layer underlying the skin is increased to cool thebody or decreased to help conserve heat. As a result, skintemperature, especially in the extremities, is a poor measureof core temperature. The skin above the temporal artery in thetemple region of the forehead provides an exception. Located just under the skin, the temporal artery provides a location inwhich the intervening tissue layer between the skin surface andan arterial blood vessel is thin. In addition, since the temporalartery branches off the carotid artery supplying the brain, its Receiver Fig. 1: Bandage based thermometer is affixed to the temporalregion of the forehead. A wireless link allows remote moni-toring of patient temperature.blood flow remains relatively constant and its temperature isclose to the core temperature.The basic operating principle of the TAT is to measure theheat flux leaving the skin in the temporal artery region alongwith the skin temperature at that point. Fourier’s law of heatconduction and the requirement for heat balance (heat fluxout of the skin must be equal to the heat flux emanating fromwithin the body) is then used to estimate the arterial bloodtemperature and hence the core temperature.To measure skin temperature, the hand-held TAT uses aninfrared (IR) imager. The user scans the IR sensor across theforehead region and the device picks the highest reading as theskin temperature over the temporal artery. In our approach, seeFigs. 1–2, an array of temperature sensors (thermistors) locatedalong the length of the bandage is affixed to the temple region.The array is electrically scanned and the highest temperaturereading is used in subsequent calculations. It should be notedthat the bandage will alter the skin temperature, but the systemaccounts for this.The second parameter that must be determined is the heatflux leaving the head in the region of the temporal artery. Oneapproach, as used in the hand-held TAT, is to estimate theheat flux from temperature difference between the skin andthe ambient air temperature. With the hand-held TAT, ambienttemperature is relatively straight forward to measure sincethe instrument is relatively large and the ambient temperaturesensor can be located sufficiently far from the forehead sensor 1-4244-0437-1/06/$20.00 ©2006 IEEE. 166  Sensor InterfaceElectronicsRF Link ElectronicsFlex PCB(Upper Portion)Flex PCB(Lower Portion)Heat Flux Foam Spacer BatteryBuried ThermistorsSkin ThermistorsTop FoamMiddle FoamBottom Foam Fig. 2: Exploded view of bandage thermometer construction.location so that its reading will be unaffected by the heatemitted by the head.For the bandage based TAT, ambient temperature measure-ment is a significant problem since the device is affixed tothe head for long periods of time. Any ambient sensor locatedon the bandage measures not the ambient room temperaturebut the micro-climate surrounding the head. The solutionto this problem is to realize that the ambient temperaturemeasurement is used only to estimate the heat flux out of the skin. By burying a temperature sensor within the bodyof the bandage TAT, as shown in Fig. 2, heat flux throughthe bandage (and hence out of the skin) can be determineddirectly. Ambient temperature in no longer needed.An important advantage of using a buried temperature sen-sor measurement for the heat flux calculation is that it dependsonly on locally measured temperatures and the physical andthermal characteristics of the bandage and not the externalenvironment. For example, if the patient is sleeping withtheir temple on a pillow, an ambient temperature based heatflux calculation would lead to an erroneous core temperatureestimation. The insulating effect of the pillow (and the bandageitself) is automatically accommodated by the new approach.II. E STIMATING  C ORE  B ODY  T EMPERATURE The temporal artery, carrying blood at core temperature, liesclose to the skin surface in the temple region of the forehead.Some of the heat transported by the blood will be lost intothe surrounding tissue and flow outward through the skin intothe environment. To simplify the discussion, we consider thetemporal artery to behave as a heat source maintained at thecore temperature of   T  c , and that the tissue layer overlyingthe artery has a conductive heat transfer coefficient of   h tissue [W /( m 2 · K  ) ]. Using Fourier’s law of heat conduction, the heatflux density flowing out of the skin,  q skin  [W / m 2 ], due to heatlost from the temporal artery, can be modelled as q skin  = h tissue ( T  c − T  s )  (1)where  T  s  is the skin temperature. The heat transfer coefficientof the tissue,  h tissue , will depend on the tissue thickness, tissuecomposition, and the level of perfusion.The heat flowing out of the skin must be balanced by theheat lost to the environment  q env , or q skin  = q env .  (2)Heat is lost from the skin into the environment by a mixtureof four mechanisms [4]–[6] q env  = q cond + q conv + q rad + q e ,  (3)consisting of   q cond , the conductive heat loss to objects incontact with the skin;  q conv , the convective heat loss due to airflowing across the skin;  q rad , the radiative heat loss; and  q e , theevaporative heat loss due to sweating. For bare skin in a coolenvironment, radiative and convective heat losses dominate.As the temperature rises, heat loss due to evaporation of sweatbecomes increasingly important.If an object is in contact with the skin, with its outer surfaceat the ambient temperature,  T  a , the conductive heat loss to theobject will be q cond  = h cond ( T  s  − T  a ),  (4)where  h cond  is the coefficient of thermal conductivity of theobject. Air flow across the skin can induce a significantconvective heat loss q conv  = h conv ( T  s  − T  a ),  (5)where the convective heat loss coefficient  h conv  is empiricallydetermined and dependent on air flow. For exposed skin,significant heat can be lost (or gained) through electromagneticradiation. Assuming that the surfaces seen by exposed skin areat a mean surface temperature of   T  a , the net radiative heat losswill be q rad  = σ( T  4 s  − T  4 a  ),  (6)where  σ  is the Stefan-Boltzmann constant (5 . 6704  × 10 − 8 J /( s · m 2 · K 4 ) ),    is the surface emissivity, and the tem-peratures are in kelvin. On the kelvin scale, the skin andambient temperatures are typically close together for indoorconditions, therefore (6) can be approximated by q r    h r  ( T  s  − T  a ).  (7)where  h r   =  σ 4 T  3 a  . The evaporative heat loss depends onthe water vapor pressure difference between the skin and theenvironment [7] q e  = h e w e ( P s − P a ).  (8)where  h e  is the coefficient of evaporation (dependent on airflow),  w e  is skin humidity [6],  P s  is the saturated water vaporpressure at the skin (depends on skin temperature), and  P a  is 167  Fig. 3: Temporal artery bandage thermometer prototype.the vapor pressure of the surrounding air (depends on relativehumidity and ambient temperature).To estimate the core temperature, the TAT must solve theheat balance equation (2). If the conductive, convective, andradiative heat transfer coefficients are amalgamated, the heatbalance equation can be rewritten as h tissue ( T  c − T  s ) = h ( T  s  − T  a ) + h e w e ( P s − P a )  (9)where  h  =  h cond  + h conv  + h r . The TAT cannot account forevaporative heat loss since the skin wetness is unknown, so afurther approximation is made h tissue ( T  c − T  s )  h ( T  s  − T  a ).  (10)From this the core temperature can be estimated using T  c   hh tissue ( T  s  − T  a ) + T  s .  (11)The apparent simplicity of (11) hides a number of issues.First, sweating can introduce significant error since theevaporative heat loss term is neglected. For the hand-held TAT,it is recommend that the temperature measurement be madebehind the ear if sweating is observed on the forehead [8],since this area has little perspiration and the arteries in thisarea will be fully dilated under these conditions.Second, the amalgamated heat transfer coefficient  h  is nota constant, but depends on air flow and temperature.Finally, the heat transfer coefficient of the tissue,  h tissue ,overlying the temporal artery depends on thickness, compo-sition, and perfusion rate. This variability is unavoidable, butthe tissue layer overlaying the temporal artery is relativelythin, consistent, and with relatively constant perfusion, therebyreducing this uncertainty.Therefore, for the hand-held TAT to make a reasonableestimate of the core temperature, a number of conditions mustbe met: no sweating, no air flow across the skin, variationof the heat transfer coefficient with temperature must beaccommodated, and the instrument must be acclimatized tothe environment immediately surrounding the forehead.By covering the temporal artery region with an insulatingbandage, the bandage TAT completely alters the significantterms in the heat balance equation. Because the skin is coveredby the bandage, radiative loss from the skin is negligible, sinceonly the outer surface of the bandage radiates. Evaporativeheat loss is also negligible since the bandage is imperviousto moisture. Convective heat loss at the skin is non-existentsince the skin is covered. Instead, conductive heat loss is thedominant term, in contrast to the hand-held TAT where it isthe only negligible term.The outside of the bandage experiences convective, ra-diative, and even evaporative heat loss (if wetted by sweatdripping on it), but this is unimportant, since the bandageTAT uses a buried temperature sensor to directly measure theheat flux out of the skin surface. Therefore the bandage TATsignificantly reduces the variability of the coefficients in theheat balance equation. The core temperature can be estimatedusing T  c   h bandage h tissue ( T  s  − T  b ) + T  s  (12)where  h bandage  is the heat transfer coefficient which is de-pendent on thermal characteristics of the bandage (well con-trolled), and  T  b  is the buried temperature within the bandage.III. P ROTOTYPE A prototype of the bandage TAT has been constructed, asshown in Fig. 3. This device replicates the key ideas of thedesign shown Fig. 2, but does not include such features as aflexible PCB for wiring.A small commercial very low power RF transceiver moduleoperating at 2.4GHz is used to provide the wireless link,and a low-power micro-controller with an 8-channel ADCprovides the sensor interface. Temperature sensing is providedby an array of four 0.1  ◦ C accurate thermistors arrangedalong the bottom surface of the bandage. Each thermistoris mounted in a 5mm diameter copper disk to increase thethermal capture area. To ease construction and to minimize thethermal shadowing caused by the skin sensor array, the arrayof three buried temperature sensors is offset from the skinsensors. The bandage is attached to the skin using adhesivetape. The entire bandage is powered by a CR2032 battery (3V,220mAh).IV. M EASUREMENTS Validation of the bandage TAT is an enormous challenge.One of the fundamental ideas of the bandage TAT is the use of the two arrays of temperature sensors to allow a direct measureof heat flow out of the temple region. This provides thebandage with the ability to adapt to changes in the environmentnear the head, for example, when a patient is sleeping withtheir head on a pillow such that the bandage TAT is covered. Toverify this, the bandage was affixed to the temple of a healthymale (with an orally measured temperature of approximately37 ◦ C). Initially the person was sitting; after a period of timehe placed his head (bandage side down) on a pillow, finallyhe returned to the sitting position. For each core temperaturecalculation, the temperature readings of two skin temperaturesensors flanking each of the buried temperature sensors wasaveraged before being applied to Eq. 12. Fig. 4 shows thecore temperature predicted by each grouping of sensors. Inpractice, the core temperature predicted by the skin sensorswith the highest average reading would be used.The core body temperature predicted by Sensor Group #2(the sensors in the middle of the bandage) fall within the rangeof 37 − 37 . 5 ◦ C. However, the absolute value of the predictedtemperature should not be considered accurate since sufficient 168  Time(hours)Sensor Group #1    B  a  n   d  a  g  e   A  p  p   l   i  e   d   B  a  n   d  a  g  e   R  e  m  o  v  e   d   B  a  n   d  a  g  e   A  p  p   l   i  e   d   B  a  n   d  a  g  e   R  e  m  o  v  e   d   B  a  n   d  a  g  e   A  p  p   l   i  e   d   B  a  n   d  a  g  e   R  e  m  o  v  e   d Sensor Group #2Sensor Group #3    T  e  m  p  e  r  a   t  u  r  e   (       ◦    C   ) 131415163032343638401314151613141516 Head on PillowHead on PillowCoreSkinBuriedCoreSkinBuriedCoreSkinBuriedHead on Pillow Fig. 4: Raw and core temperature measurements. Each calculated core temperature is based on the average readings of twoadjacent skin temperature sensors and the intervening buried temperature sensor. Initially the person is in the sitting position,at time of approximately 13.5 hours, the person places their head on a pillow (bandage TAT is covered by the pillow) andremains in that position until approximately 14.25 hours, when the person sits up has not yet been gathered to allow the factor  h bandage h tissue inEq. 12 to be properly calibrated. What is more important inthese results is the minimal variation in core temperature thatis observed between the subject in the sitting and head onpillow positions.Observe the significant increase in skin temperaturerecorded when the head is placed on the pillow. This isexpected as the pillow insulates the head. Consider the middleset of sensor data shown in Fig. 4. Except for a transient, thepredicted core temperature remains relatively constant betweenthe sitting and head on pillow positions, even though theskin temperature has significantly changed between these twopositions. As the skin temperature rises during the head onpillow time, so does the buried temperature, but with a decreas-ing difference. This reflects the reduced heat flow out of thetemple region due to the insulating effect of the pillow, clearlydemonstrating the ability of the technique to accommodatelocal variation in the environment. After the subject returnsto the sitting position, the skin and buried temperatures drop,and a small increase in predicted core temperature is observed.We believe this occurs due to the improved thermal contactbetween the skin temperature sensors that results from thepressure applied to the bandage while the head was on thepillow. Ensuring consistent thermal contact of the sensors withthe skin is one of the on-going design challenges for thisproject. The two outer sensor groups show a larger variationin predicted core temperature between the sitting and head onpillow positions. One possible explanation for this is that inthe prototype used, the two PCBs for the wireless link andthe sensor interface circuitry overlay these two other sensorgroupings and alter the thermal characteristics of the bandagein these regions.The results demonstrate that the fundamental operatingprinciple of the bandage TAT appears promising. More ex-tensive testing with patients presenting with a range of coretemperatures is required to validate the clinical accuracy of the device. A next generation bandage TAT is in developmentthat will be suitable for limited clinical trials.V. C ONCLUSIONS A bandage based Temporal Artery Thermometer has beenconstructed that provides a means of wirelessly monitoringa patient’s core body temperature in a continuous and non-invasive fashion. A heat-flux sensor system incorporated intothe bandage makes it less sensitive to errors induced byperspiration, convection, and radiative heat loss than a standardhand-held Temporal Artery Thermometer.R EFERENCES[1] F. Pompei, “Temporal artery temperature detector,” U.S. Patent 6292685B1, Sept. 18, 2001.[2] ——, “Ambient and perfusion normalized temperature detector,” U.S.Patent 6499877 B2, Dec. 31, 2002.[3] F. Pompei and M. Pompei, “Non-invasive temporal artery thermometry:Physics, physiology, and clinical accuracy,” in  Proceedings of SPIE  , M. R.Dury, E. T. Theocharous, N. J. Harrison, M. Hilton, and N. Fox, Eds.,vol. 5405, Apr. 2004, pp. 61–67.[4] D. Fiala, K. J. Lomas, and M. Stohrer, “A computer model of humanthermoregulation for a wide range of environmental conditions: thepassive system.”  J Appl Physiol , vol. 87, no. 8750-7587, pp. 1957–72,1999.[5] Y. H. Chiok, E. Y.-K. Ng, and V. V. Kulish, “Global bioheat modelfor quick evaluation of the human physiological thermal profiles underdiffering conditions.”  J Med Eng Technol , vol. 26, no. 0309-1902, pp.231–8, 2002.[6] Z.-S. Deng and J. Liu, “Mathematical modeling of temperature mappingover skin surface and its implementation in thermal disease diagnostics.” Comput Biol Med  , vol. 34, no. 0010-4825, pp. 495–521, 2004.[7] S. B. Wilson and V. A. Spence, “A tissue heat transfer model for relatingdynamic skin temperature changes to physiological parameters.”  Phys Med Biol , vol. 33, no. 0031-9155, pp. 895–912, 1988.[8] F. Pompei and M. A. Pompei, “Temporal thermometer disposable cap,”U.S. Patent 6,932,775 B2, Aug. 23, 2005. 169
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks