|1||Infrared Thermometers||Is there any way to validate body site offset correction at NMIs?
If they are within laboratory accuracy, will they give correct body temperature?
|Yuan Zundong*||The correction of body site bias is a statistical estimate of body site temperature difference of a group of people with a distribution under certain conditions. Therefore, the correction can’t be verified using the standards of laboratory measuring instruments only.
Different forehead thermometers and screening thermographs are likely to adopt different correction models.
The laboratory accuracy of an infrared thermometer reflects its accuracy of radiation temperature measurement, but not the accuracy of its correction or its statistical estimate for the body temperature. Moreover, a proper correction, i.e. a statistically best estimate, does not mean that it is very close to the correction for a specified person. The clinical bias reflects how accurate the statistical estimate for the body temperature is.
|2||Blackbodies for Calibration||Can we calibrate without a black body source?||Yuan Zundong||Infrared thermometers for human body temperature measurements can’t be calibrated without using black body sources. Infrared thermometers which indicate body temperature are designed to be calibrated in calibration mode (unadjusted mode, or test mode) in which the thermometer indicates the radiation temperature (the black body temperature).|
|3||Infrared Thermometers||The principle of infrared forehead thermometers measuring body temperature is a combination of detecting the forehead surface temperature and applying a surface-body temperature transformation model. However, there is no uniform transformation model, which induces that we cannot confirm the calibrated infrared forehead thermometer using a black body can be used to measure the body temperature accurately.
How to solve this problem? To establish a uniform forehead surface-body temperature transformation model or calibrate the infrared thermometer in body temperature measurement model?
|Yuan Zundong||Different correction models result in different body temperature estimations, but a uniform (universal) correction model has to be applicable to a wide range of measurement conditions and a variety of people. The randomness of measurement conditions and subjects would lead to insufferable difference between their actual and statistical estimated body temperatures. In principle, possible solutions for forehead thermometer applications include: to reduce the dispersion of measurement conditions and subjects, or to make the difference between the measurement conditions and/or between the subject's characteristics to be known in advance, or to choose a site with a sufficient and stronger correlation between skin and body temperatures, for example, inner canthus but better for thermograph.|
|4||Thermal Imagers||Has there been any correlation between inner canthi to core body temperature?||Wang Li||As far as I know, there is not an established correlation by clinical study. It is studied by thermal imager suppliers and users individually|
|5||Infrared Thermometers||Since the display does not indicate the body temperature, if we calibrate an IR Thermometer using the non-adjusted mode, how can we get clinical accuracy?||Yuan Zundong||The accuracy of infrared thermometers indicating body temperature consists of two parts, laboratory accuracy (metrological accuracy of thermal radiation or radiation temperature measurement), and clinical accuracy (medical application accuracy of body temperature measurement). The former is determined under calibration mode, while the latter requires the comparison of the IR measurement results of a group selected people under body temperature mode and under specified conditions and the measurement results of body core temperatures.|
|6||Blackbodies for Calibration||What is the recalibration interval of a black body source?||Lu Xiaofeng||There is no any restrict recalibration interval for a black body source. Two years are a good choice for ear and forehead thermometer blackbodies from our experience.|
|7||Blackbodies for Calibration||Which calibrator do you recommend for body IR thermometers, say, fluke 4180?||Lu Xiaofeng||Fluke 4180 is not recommended for a standard infrared thermometer due to its uniformity and emissivity. A standard ASTM or EN or JST cavity with a painting immersed in a bath should be recommended.|
|8||Blackbodies for Calibration||Why the cavities dedicated to IRET and IRFT are different considering that both thermometers integrate the flux over a full hemisphere?||Lu Xiaofeng||Different aperture of cavity is needed for the calibration of IRET and IRFT. Large aperture needs long deepth for the same emissivity.|
|9||Infrared Forehead Thermometers||Don't we need to perform any distance dependent measurement for the forehead thermometer eg. the range given with the thermometer?||Lu Xiaofeng||The thermometer should always be used in the distance suggested by the producer. But a good thermometer could be independent of the measuring distance within the specified range.|
|10||Blackbodies for Calibration||What is the effectiveness of infrared to measure temperature of a black body in Lauda oil?||Lu Xiaofeng||A bath can help the black body cavity have a good temperature stability and uniformity which means the emissivity will be more close to the calculation.|
|11||Infrared Thermometers||What is the practical approximation for the emissivity of ears and foreheads?||Lu Xiaofeng||The emissivity of ear canal could be estimated to be 0.99 and the forehead skin is 0.97~0.98.|
|12||Blackbodies for Calibration||Is the similar calibration source used to perform the calibration of the forehead thermometer and ear thermometer?||Wang Li||Yes, it is. But some forehead thermometers have distance detectors to ensure the thermometers are used at the requested distance. For calibration of such thermometers the traditional cavity type of black body calibration sources are not suitable as the distance detectors would not be triggered and the measurements would not be activated.|
|13||Thermal Imagers||What system meets all these requirements?||Wang Li||It's hard to say any without proper tests.|
|14||Thermal Imagers||Regarding evaluation/selection of thermal imagers document standard IEC80601-2-59:2017, speaks of using the MRTD (ASTM E1213-14) rather than ASTM1311-14 on MDTD. Which ASTM standard should I refer to?||Wang Li||In my opinion, no one is suitable for evaluation of thermal imager for body temperature screening. That's why we define MDTD in SS582 differently.|
|15||Thermal Imagers||If a person comes from outside in a very cold weather, does he/she need som time to thermally stablise before thermal image taking for pre-screening the fever? Is it reliable to use thermal imagers outside in a very cold/hot whether?||Wang Li||Answered during Q&A of seminar.
Yes, the person needs to have sufficient time to get thermally stable when he/she comes from cold outside.
From the thermal imager point of view, as long as it's within its required ambient operation conditions, besides some technical concerns such as temperature coefficient, there is not a big issue. However, the problems come from human being. The skin temperature is affected by the cold/hot whether and that makes the screening not reliable.
|16||Thermal Imagers||Thermal Imagers for Human Temperature Screening :
Question 1: The parameters of the infrared thermal imagers: temperature range (30~50)℃, emissivity 0.98 (approximate human skin emissivity), MPE: ±0.4℃, infrared thermal imagers with error correction function and high temperature warning function, without body temperature function, can they be used for body temperature screening?
|Wang Li||There are many parameters need to be considered beyond the few mentioned. Thermal Imagers for human temperature screening is a holistic approach.|
|17||Infrared Thermometers||Can you use infrared thermometers on humans and are they accurate?||Lu Xiaofeng||The industrial infrared thermometers cannot be used on body temperature measurement as their uncertainties are always larger than 1 ℃.|
|18||Infrared Thermometers||What is the difference between Forehead IR Thermometer and Ear IR Thermometer?||Lu Xiaofeng||IRFT measures the skin temperature which could be different from the body core temperature. And the skin emissivity is not always the same for different people and in different situations. IRET can measure the ear canal temperature which is more close to the body core temperature, and the emissivity of ear canal is also close to 1.|
|19||Infrared Forehead Thermometers||How to estimate the body-site offset correction when doing the IRFT calibration?||Bai Chengyu||Ususally we contact manufactures to inquiry. Correction may also be determined by experiments when we know how to set the IRFT into calibration mode. One point we should notice is that the correction may vary with ambient temperature.|
|20||Infrared Forehead Thermometers||Are the region sizes for IRFTs similar?||Bai Chengyu||Target size required by IRFTs may be quite different between models. Refer to Page 13-14 of Bai Chengyu's presentation.|
|21||Infrared Forehead Thermometers||What is the highest MU component when calibrating IRFT?||Bai Chengyu||The standard uncertianty of BB is usually 0.02℃. The contributions from IRFT are usually repeatability, resolution and measurment distance effect.|
|22||Infrared Forehead Thermometers||Is there a possibility of having adjustments options on IR thermometers for offsets for calibaration purposes?||Bai Chengyu||We found some IRFTs sold in China have such a function. Users can set Offset by botton operation.|
|23||Infrared Forehead Thermometers||What mode is an IRFT calibrated in at NIM China, surface mode or body mode?||Bai Chengyu||Calibration shall be carried out in calibration mode. Some IRFTs may use "surface" to indicate calibration mode. But please note that some multi-use IRFTs may adopt "surface mode" to measure object temperatrue, in which the target’s emmissivity is set to 0.95.|
|24||Infrared Thermometers||What is the calibration period for IR Clinical Thermometers? Is this recommended or required within Europe?||Bai Chengyu||Calibration period for clinical thermoemters is one year in China, and life-cycle is four years.|
|25||Infrared Thermometers||What calibration points to use? Are they set by the laboratory or the customer? What is the best practice to perform calibration? Is it from the highest point to the lowest or from the lowest to the highest?||Bai Chengyu||We usually calibrate IRETs at 35℃, 37℃ and 41.5℃. The tempearature points can be changed in calibration when the customer requests. The sequence of calibration points is not regulated.|
|26||Infrared Thermometers||In India, I have observed use of IRFT generally in the COVID-19 coronavirus pandemic. So is IRET more expensive than IRFT? What is the cost comparison between the two?||Bai Chengyu||The prices of IRETs and IRFTs are similar. But the usage cost of IRETs may be higher than IRFTs, because disposable probe covers shall be adoped for public use.|
|27||Infrared Forehead Thermometers||There are 2 settings for IR clinical thermometer: surface setting and body setting. When performing the infrared thermometer calibration, do we set to "surface" for our test setting?||Bai Chengyu||Calibration shall be carried out in calibration mode. Some IRFTs may use "surface" to indicate calibration mode. But please note that some multi-use IRFTs may adopt "surface mode" to measure object temperatrue, in which the target‘s emmissivity is set to 0.95, please contact mannufactures to confirm.|
|28||Infrared Forehead Thermometers||What is the right distance for one to measure human temperature "forehead" using an infrared thermometer?||Bai Chengyu||Please follow the user's mannual instructions. Refer to Page 12-15 of Bai Chengyu's presentation .|
|29||Infrared Thermometers||Is it true that most of ear and forehead thermometrers have fixed emissivity? It is crucial to know what is the built up/fixed emissivity set by the manufacturers in order to calculate radiance temperature of black body radiator. I use this radiance temperature as reference temperature.||Bai Chengyu||IRETs and IRFTs shall be set into calibration mode (or test mode) when doing calbration, in which the object emissivity is preset to 1.0.|
|30||Blackbodies for Calibration||The black body is used as a standard for calibration of IR thermometers. Is the black body for calibration of clinical IR thermometers different in design or metrological principles from the black body for industrial IR thermometer calibrations? Any specific requirements seriously for clinical IR Thermometer calibrations compared with industrial IR Thermometer calibrations?||Lu Xiaofeng||Yes, they are totally different from industrial IR thermometer calibrations. The uncertainty of the standard black body is as small as 0.07℃(k">=2), which means the uniformity and stability of the black body cavity should be on the order of dozens of mK.|
|31||Infrared Thermometers||How to do mass calibration/verification of IR thermometers currently used for screening COVID-19 patients?||Bai Chengyu||We may reduce number of temperature points to cope with mass calibrations in emergency. For example, 37℃ for IRETs and 35℃ for IRFTs.|
|32||Infrared Thermometers||How to calibrate IR thermometers for human skin without knowing the algorithm used or instrumental emissivity?||Yuan Zundong||Infrared thermometers for human skin should be calibrated using a black body source in calibration mode (non-adjustment mode, or test mode). In this mode, the thermometer indicates the measured radiation temperature (black body temperature). Calibration does not involve human-related characteristics.|
|33||Blackbodies for Calibration||The best practice for calibrating infrared thermometers is to use dry block calibrators with black body inlet, right?||Lu Xiaofeng||No, the best choice for calibration of infrared thermometers is a bath with black body cavity.|
|34||Infrared Thermometers||Can you provide instructions/suggestions on calibration and verification laboratories? How can we verify the clinical bias at laboratory level?||Yuan Zundong||Clinical bias should be validated under measurement conditions with subjects with some distribution that are close to those specified by the thermometer manufacturer. It can also be determined under the specific distributed subjects and specific measurement conditions required. The clinical bias is the mean of the differences between the body temperatures indicated by the infrared thermometer of these subjects and their body core temperatures. Therefore, clinical bias can’t be determined in the laboratory using a black body source. However, a thermometer with a qualified laboratory accuracy is the premise to determine clinical bias.|
|35||Infrared Thermometers||Can you provide instructions/suggestions on calibration and verification laboratories? How to calibrate infrared thermometer?||Bai Chengyu||a) make sure the optics is clean and clean up when needed.
b) keep IRFTs and IRETs in a stable environment for 30 minutes or longer as manufactures claim. Please keep the probe covers at the same environment if applied.
c) set the IRETs and IRFTs into calibration mode.
d) keep time intervals as manufactures require when measuring a black body.
e) use probe covers designed by manufactures (for IRET)
f) take measurement at minimum and maximum distance claimed by manufactures (for IRFTs)
|36||Infrared Thermometers||How often do you recalibrate an infrared thermometer?||Bai Chengyu||Caibration period for clinical thermoemters is one year in China, and the life-cycle is four years.|
|37||Blackbodies for Calibration||Optimal equipment for calibration of clinical infrared thermometers?||Lu Xiaofeng||A standard ASTM or EN or JST cavity with a painting immersed into a bath is recommended.|
|38||Blackbodies for Calibration||Could you suggest the most suitable black body to calibrate IR (forehead) thermometer?||Lu Xiaofeng||A standard ASTM or EN or JST cavity with a painting immersed into a bath is recommended.|
|39||Blackbodies for Calibration||Is a black body suitable to calibrate all IR Thermometer brands?||Lu Xiaofeng||An ear thermometer black body is suitable for calibration of all ear thermometers but is not suitable for calibration of forehead thermometers.|
|40||Blackbodies for Calibration||What is the best reference instrument to calibrate IR Thermometers?||Lu Xiaofeng||A standard ASTM or EN or JST cavity with a painting immersed into a bath is recommended. NIM has supplied dozens of reference insturments to calibration labs.|
|41||Blackbodies for Calibration||How do we define the suitable emissivity for measuring the human body temperature? What if I only have a black body with e=0,95?||Lu Xiaofeng||The emissivity of the ear canal can estimated as 0.99 and the forehead skin as 0.97~0.98. The black body with e=0.95 cannot be used as a standard for the calibration of infrared clinical thermometers.|
|42||Application of Infrared Thermometers||I would like to know your experience regarding body temperature measurements during the pandemic.||Bai Chengyu||Please refer to Page16 of Bai Chengyu's presentation .|
|43||Application of Infrared Thermometers||Are thermometers influenced by the environment?||Bai Chengyu||Yes. For example, radation from the envirement affects IRFTs obviously.|
|44||Infrared Forehead Thermometers||Most forehead thermometers display the measurement result as an estimated oral temperature. How to deal with this situation?||Bai Chengyu||In the metrology lab, we calibrate IRFTs in calibraion mode.|
|45||Infrared Ear Thermometers||How can I set the emissivity for a thermometer if I want to measure the ear temperature more accurately?||Bai Chengyu||Sorry, we haven't seen any ear thermometer which has an emissivity setting function.|
|46||Infrared Thermometers||What is the impact of skin type on emissivity?||Yuan Zundong||The skin emissivity is around 0.95 in 3~5 μm and around 0.98 in 8~14μm in early reports, and the effect of skin type on skin emissivity is rarely reported. However, different skin tones have not been found to have a significant effect on emissivity, much like both white and black paints with high emissivity at 8~14 μm. Note that skin will reduce its temperature and emissivity after sweating or being applied makeup.|
|47||Application of Infrared Thermometers||How is the industry adopting the NMI calibration of IR Thermometers?||Graham Machin||In my experience there are good traceability chains to accredited laboratories for industrial IR thermometers in the Metrology Regions.
This is less so for thermal imagers – companies do recognise the need for calibration but often that calibration is not traceable to national standards.
For medical IR thermometers the traceability is not clear it depends upon country and local standards and regulations. Certainly in my experience clinical thermometers even when calibrated are not always traceably calibrated.
|48||Blackbodies and calibration||What is the required uncertainty level for calibration of IR thermometer used for clinical purposes?||Lu Xiaofeng||The uncertainty of a calibrated clinical IR thermometer is usually required to be lower than 0.2℃(k=2).|
|49||Blackbodies and calibration||Which factors need to be considered in estimation of measurement uncertainty? How to estimate practically?||Lu Xiaofeng||The factors include: the uniformaity and stability of the bath, the SPRT and resistance bridge, emissivity and repeatability.|
|50||Blackbodies and calibration||Accuracy of measuring a tem of a liquid using IR-TM (near-melting, boiling, middle points)||Lu Xiaofeng||It depends on the emissivity estimate of the liquid.|
|51||Blackbodies and calibration||How to develop the uncertainty budget for infrared thermometer calibration?||Lu Xiaofeng||Please refer to the reference: Uncertainty Budgets for Calibration of Radiation Thermometers below the Silver Point.|
|52||Infrared Thermometers||Can you talk about reliability of IR measurement, effect of environment, uncertainty components and standard correlation between body & surface mode?||Yuan Zundong||For the use of ear thermometers, the biggest risk is operational skills or subject’s excessive earwax. For the use of forehead thermometers, the biggest “measurement” risk is from its measurement principle constraints. The ear thermometer measures a body core temperature, and the forehead thermometer measures the forehead skin radiation temperature and estimates the body core temperature using a statistical correction model. For the measurement of the body core temperature using a forehead thermometer, the forehead temperatures of a group of subjects under a set of measurement conditions are highly dispersed around the mean. So the statistical estimate for the body temperature is not necessarily the actual body temperature measurement of an individual subject.|
|53||Other||Is the new TG of CCT "Body Temp Meas" only related to IR thermometers or it can propose new techniques, e.g. using SWIR imaging?||Graham Machin||The TG is presently restricted to all forms of IR body temperature measurement. This includes: ear, forehead/temporal artery and all forms of thermal imaging, including SWIR.|
|54||Other||How can those National Metrology Institutes with less laboratory capacity be able to implement these skills after training?||Klaus Anhalt||NMIs starting in the field of radiation thermometry are probably not new to thermometry in general, i.e. they are familiar with ITS-90, temperature fixed-points, contact temperature sensors, calibration baths or dry well calibrators. Some instruments used for calibration in contact thermometry can be used to build a reference blackbody source for radiation thermometers, for example by using commercially available blackbody insets for water bath or drywell calibrators. With a radiation thermometer pointing at this radiation source, such a setup can be used to investigate certain measurement effects in radiation thermometry or even calibrate the radiation thermometer. The following document discusses the calibration of a radiation thermometer using a cavity radiator immersed in a water bath https://www.ptb.de/cms/fileadmin/internet/publikationen/wissensch_tech_publikationen/Testing_the_suitability_of_cavity_radiators_used_for_the_metrological_verification_of_infrared_thermometers__Supplement_to_LMKM_.pdfPTB collected links, documents and guidelines for the calibration of radiation thermometer with special focus on medical non-contact thermometers such as ear thermometershttps://www.covid19.ptb.de/calibration-of-infrared-thermometers|
*Note: Dr. Yuan Zundong is a NIM Fellow and senior researcher at NIM whose research interests cover radiation thermometry etc. His email is firstname.lastname@example.org.