Infrared thermometers - good/crap/other options?
Infrared thermometers - good/crap/other options?
Author
Discussion

g3org3y

Original Poster:

22,462 posts

220 months

Thursday 21st May 2020
quotequote all
A bit of advice regarding infrared thermometers if anyone has any experience please.

We've split the GP surgery into two sections: Red Zone for suspected Covid patients, Green Zone for all other patients, nursing stuff etc.

For the Green Zone, the plan was to use an infrared thermometer to check the temperature of all patients coming in for an appointment. If a patient has a temp of 37.8 or above, they are not allowed into the Green Zone and may need to be seen in the Red Zone. (Let's set aside the issue of asymptomatic covid positive patients for the time being...worms/can/open etc)

We do however have a few issues:
1) Thermometer has to be used at a distance of 3-5cm from the forehead meaning the staff member is breaking the 2m social distancing 'rule'.
2) Thermometer seems to under read to a variable extent.

Staff members have expressed concern and worry about breaking the 2m rule and are consequently reluctant to check the temps. I should add staff are wearing masks + gloves + surgical scrubs (so appropriate protection for the role imo).

Yesterday we had an issue where the infrared thermometer recorded a normal temp at the door but subsequently when the clinician checked the temp during the clinical examination using an proper ear thermometer it was raised.

How accurate are these infrared thermometers? We bought one from a reputable source ( proper medical supplier) so it's not a crap cheapo one. I'm aware that user operation can affect things (perspiration, strand of hair, ambient temp etc).

We don't want patients to do their own temperature at the door as we don't want them handling the equipment (infection risk).
We don't want to do ear temps at the door as this requires actual patient contact (again infection risk).

Are there better 'gold standard' IR thermometers out there (that work from a 2 metre distance)?

I looked at thermal imaging cameras as another option but they are rather expensive.

Appreciate any experiences or suggestions. TIA. smile

untakenname

5,351 posts

221 months

Thursday 21st May 2020
quotequote all
I've got a cheap £10 for trackdays to measure tyre temperature and it's accurate and gives the same reading when compared to traditional digital thermometer pressed against the tyre though neither have been calibrated so unsure of the true accuracy.

Perhaps a thermometer that was mounted on a tripod with a remote trigger or delay would solve the 2 meter issue?

Interesting article about using infrared thermometers and coronavirus here

https://www.nytimes.com/2020/02/14/business/corona...

Phud

1,479 posts

172 months

Thursday 21st May 2020
quotequote all
I use a fluke, easy works and you can buy then with test certs to prove their accuracy

STIfree

1,913 posts

188 months

Thursday 21st May 2020
quotequote all
Contact the medical/clinical engineering department who maintain the rest of your medical equipment.

As you're a surgery I'd imagine your part of a larger trust who will have their own engineering department.

They will give you the best advice on what certified equipment that you should use.

andrewjamesroberts

2,266 posts

233 months

Thursday 21st May 2020
quotequote all
Head to the pizza oven thread!

There will be loads of advice in there!

anonymous-user

83 months

Thursday 21st May 2020
quotequote all
Have pinged this to a mate who may be able to help

jet_noise

6,090 posts

211 months

Thursday 21st May 2020
quotequote all


Eek, we're being watched wink

Possible grandmother egg sucking info;
Thermal imaging cameras are IR devices. Just a lot of them in a panel! IR is the measuring technique.

How do you sight the medical devices and how large is the sensing area at the recommended distance? i.e how do you know they're pointing at the (right part) of a person's forehead? Just by being close?
I imagine using a thermal imaging camera or from further away might be more tricky to get that right.
Ones I've used can provide a direct readout of display centre and see in the display that it was over the desired hot (or cold) item.
Or you can turn on a sighting beam illuminating the target. I think this might be a laser pointer type thing (I never used it, always sighted by, er, sight!). Don't think I'd want to point that at a person!

I used thermal image cameras over some years for thermal analysis of power electronic circuit boards, truck diesel ECU, with many moderately chunky semiconductors. Checking dissipation calcs. and that thermal paste to the casting was in the right places.

On the staff worry I don't think you can do anything but point at the guidelines. For MoPs isn't the guidance 2m without PPE. PPE is only recommended when closer. There's an excellent video from NHS Prof Kiernan(?) on YT that might help get the message over. If still current.
One other addition might be a see through shield, bit like a very lightweight welders mask, eyes can also be a route for droplets to get in.

IANAD just an opinionated PH poster smile

Top Banana

459 posts

241 months

Thursday 21st May 2020
quotequote all
I'm a product manager for a company that makes various temperature measuring instruments for industry and other applications - there has been a massive lift in enquiries about IR thermometers and full thermal imaging camera systems (again which we manufacture and supply).

For IR thermometers to be used with any degree of accuracy (and the same with Thermal imaging cameras) there are several variables that need to be right - firstly they need to be set with the emissivity value of the material you are measuring (all objects emit infra red energy but at differing emissivity values)

next there is the issue of spot measurement size - this is the size of the area that the detector in the IR thermometer is measuring - this is based on the optics and most thermometers will quote a spot ratio of something like 10:1, or 12:1. you need to usually look through the manufacturers info on this, but ideally you want a device with the largest ratio you can get for your money as this will technically have a smaller
spot area size to measure (this helps general accuracy as you are then only measuring the object you want and not picking up general IR in the background) as an example we have a unit with 50:1 optics which is being used in healthcare applications currently.

for units that use lasers to mark/identify the spot size, be aware that using these to monitor forehead temperature you also need to take into account the issue of making sure that people have eyes shut as you dont want to be shining laser light into peoples eyes !

Then you have issues such as making sure somebody operating the device is using it at exactly 90 degrees angle to the surface otherwise you can again get errors.

Even with everything done to make measurement as good as possible, you still will only get an accuracy typically of plus/minus 1 Deg.C - a contact surface measurement will always be more accurate, but the benefit of IR thermometers is the non-contact aspect 9in industry you can monitor moving equipment as an example)

There are thermal imaging cameras with specialist features that allow some form of automatic detection of increased body temperature (and hence fevers..) but these are highly specialised devices and expensive (again we manufacture such products but you are looking at 12k cost per unit!)

if you wanted more advice off line PM me and I can give you more guidance

Regards - T.b.

Mr Pointy

13,359 posts

188 months

Thursday 21st May 2020
quotequote all
I'm a bit puzzled: nowhere on the page with the specifications for that device does it mention how accurate it is. Surely that's the most critical piece of information you'd look for before buying it?

Given your staff have all the PPE required why not just give them the ear thermometers & tell them to get on with it?

Benrad

656 posts

178 months

Thursday 21st May 2020
quotequote all
How repeatable do you need it to be? What was the difference between the reading with the IR and reading with the in ear?

Maybe try some practise between staff? Take someone's temperature in ear, then take 100 IR readings of the same person by different members of staff and have a look at how repeatable it is. Try different things like whether the patient has sunlight on their face at the time, whether they're sweating or their skin is dry to see how that impacts repeatability

Top Banana

459 posts

241 months

Thursday 21st May 2020
quotequote all
i have just taken a look at the unit you provided the link for.

that unit looks very 'generic' and has most likely come from the far east, as per my post above being in this industry I am seeing a lot of
products touted for non-contact temperature screening but many are giving false promises.

Straight away I would question where are the details on spot size / optics ratio, and also there are NO specific accuracy levels quoted for the device. Also no sign of ability to adjust emissivity level for subject (it could have default to human skin which is 0.98 but it would be great if they told you this)

Also if you point this at someone who has just run in from outside or been standing in the sun, the readings will be high - ideally you need to allow the subject to acclimatise before taking a reading (and in addition have taken a 'baseline' reading against persons who are known free from fever and have a stable body temperature)

The offer from above is there - if you want to discuss more offline just PM me

Mastodon2

14,324 posts

194 months

Thursday 21st May 2020
quotequote all
Is the IR thermometer you're using a certified medical device designed with the express purpose of measuring temperature in human patients? Does it have a CE mark? Is there evidence that it has been calibrated to any particular standard of accuracy?

If you've got a generic IR thermometer which isnt specifically intended for use in a medical setting then I would feel cautious about putting too much stock in the readings it gives.

g3org3y

Original Poster:

22,462 posts

220 months

Thursday 21st May 2020
quotequote all
untakenname said:
I've got a cheap £10 for trackdays to measure tyre temperature and it's accurate and gives the same reading when compared to traditional digital thermometer pressed against the tyre though neither have been calibrated so unsure of the true accuracy.

Perhaps a thermometer that was mounted on a tripod with a remote trigger or delay would solve the 2 meter issue?

Interesting article about using infrared thermometers and coronavirus here

https://www.nytimes.com/2020/02/14/business/corona...
I was actually reading that article earlier.

Accuracy is our concern. A bad test is worse than no test. If it over reads (like a speedo), we're extra cautious so that's not too bad. If it under reads, we're in trouble.

I'm not sure about a remote trigger.

Ours is a handheld like this:


Phud said:
I use a fluke, easy works and you can buy then with test certs to prove their accuracy
Never heard of a fluke. Will check it out. We really want a non touch method.

STIfree said:
Contact the medical/clinical engineering department who maintain the rest of your medical equipment.

As you're a surgery I'd imagine your part of a larger trust who will have their own engineering department.

They will give you the best advice on what certified equipment that you should use.
None of that. Each GP surgery, although under a CCG umbrella is very much independent. As such, left to own devices.

I have no idea if temperatures are being checked at the front door for any patient going into hospital (outpatient clinics etc).

andrewjamesroberts said:
Head to the pizza oven thread!

There will be loads of advice in there!
biggrin

anonymous said:
[redacted]
thumbup

jet_noise said:
Possible grandmother egg sucking info;
Thermal imaging cameras are IR devices. Just a lot of them in a panel! IR is the measuring technique.

How do you sight the medical devices and how large is the sensing area at the recommended distance? i.e how do you know they're pointing at the (right part) of a person's forehead? Just by being close?
I imagine using a thermal imaging camera or from further away might be more tricky to get that right.
Ones I've used can provide a direct readout of display centre and see in the display that it was over the desired hot (or cold) item.
Or you can turn on a sighting beam illuminating the target. I think this might be a laser pointer type thing (I never used it, always sighted by, er, sight!). Don't think I'd want to point that at a person!

I used thermal image cameras over some years for thermal analysis of power electronic circuit boards, truck diesel ECU, with many moderately chunky semiconductors. Checking dissipation calcs. and that thermal paste to the casting was in the right places.

On the staff worry I don't think you can do anything but point at the guidelines. For MoPs isn't the guidance 2m without PPE. PPE is only recommended when closer. There's an excellent video from NHS Prof Kiernan(?) on YT that might help get the message over. If still current.
One other addition might be a see through shield, bit like a very lightweight welders mask, eyes can also be a route for droplets to get in.

IANAD just an opinionated PH poster smile
Exactly the problem we've got. There's no visitble beam from the thermometer so the operator can't gauge exactly where it's firing. Plus they are probably all holding it at different distances.

The other issue I've noticed with IR thermometers is that some are reportedly not suitable for human (or animal) temp measurement apparently.

With the thermal camera, not sure what the field of view is and whether we could have it mounted on a tripod from a distance and have someone to monitor. If anyone comes up 'hot', we actively check their temp with the (accurate) ear thermometer.

Unfortunately guidelines seem to go out the window when it comes to things like this. Once one staff member feels there's a risk all others follow suit. All staff have been properly risk assessed. All vulnerable staff have no patient contact at all (even if it's across a desk >2m away). As mentioned, they all have surgical masks and gloves. We have visors as well if needed.

Edited by g3org3y on Thursday 21st May 16:41

g3org3y

Original Poster:

22,462 posts

220 months

Thursday 21st May 2020
quotequote all
Top Banana said:
I'm a product manager for a company that makes various temperature measuring instruments for industry and other applications - there has been a massive lift in enquiries about IR thermometers and full thermal imaging camera systems (again which we manufacture and supply).

For IR thermometers to be used with any degree of accuracy (and the same with Thermal imaging cameras) there are several variables that need to be right - firstly they need to be set with the emissivity value of the material you are measuring (all objects emit infra red energy but at differing emissivity values)

next there is the issue of spot measurement size - this is the size of the area that the detector in the IR thermometer is measuring - this is based on the optics and most thermometers will quote a spot ratio of something like 10:1, or 12:1. you need to usually look through the manufacturers info on this, but ideally you want a device with the largest ratio you can get for your money as this will technically have a smaller
spot area size to measure (this helps general accuracy as you are then only measuring the object you want and not picking up general IR in the background) as an example we have a unit with 50:1 optics which is being used in healthcare applications currently.

for units that use lasers to mark/identify the spot size, be aware that using these to monitor forehead temperature you also need to take into account the issue of making sure that people have eyes shut as you dont want to be shining laser light into peoples eyes !

Then you have issues such as making sure somebody operating the device is using it at exactly 90 degrees angle to the surface otherwise you can again get errors.

Even with everything done to make measurement as good as possible, you still will only get an accuracy typically of plus/minus 1 Deg.C - a contact surface measurement will always be more accurate, but the benefit of IR thermometers is the non-contact aspect 9in industry you can monitor moving equipment as an example)

There are thermal imaging cameras with specialist features that allow some form of automatic detection of increased body temperature (and hence fevers..) but these are highly specialised devices and expensive (again we manufacture such products but you are looking at 12k cost per unit!)

if you wanted more advice off line PM me and I can give you more guidance

Regards - T.b.
Thank you for that very informative post.

Mr Pointy said:
I'm a bit puzzled: nowhere on the page with the specifications for that device does it mention how accurate it is. Surely that's the most critical piece of information you'd look for before buying it?

Given your staff have all the PPE required why not just give them the ear thermometers & tell them to get on with it?
I don't think it's that simple tbh. These aren't clinical staff (HCAs or Nurses) just receptionists. I know other surgeries have their HCAs on the front door doing this.

Top Banana said:
i have just taken a look at the unit you provided the link for.

that unit looks very 'generic' and has most likely come from the far east, as per my post above being in this industry I am seeing a lot of
products touted for non-contact temperature screening but many are giving false promises.

Straight away I would question where are the details on spot size / optics ratio, and also there are NO specific accuracy levels quoted for the device. Also no sign of ability to adjust emissivity level for subject (it could have default to human skin which is 0.98 but it would be great if they told you this)

Also if you point this at someone who has just run in from outside or been standing in the sun, the readings will be high - ideally you need to allow the subject to acclimatise before taking a reading (and in addition have taken a 'baseline' reading against persons who are known free from fever and have a stable body temperature)

The offer from above is there - if you want to discuss more offline just PM me
You're right, is made in China (Guangzhou Berrcom). I think the belief was, if from reputable supplier, would be a decent product. It is CE marked (0123).

I'll send a PM your way.

Edited by g3org3y on Thursday 21st May 17:14

jet_noise

6,090 posts

211 months

Thursday 21st May 2020
quotequote all
Duh, visor. That's the word for a see through plastic shield of course smile

I would say it is possible to use a thermal camera to remotely monitor from 2m away. Tb would seem to be in the business and be able to confirm. And how much!
An ideal piece of kit would be able to indicate automatically just from someone approaching without human intervention. But that also assumes patients would cooperate!
Is even the stuff used at airports (say) capable of this or is here always human aiming/interpreting?

Lastly is there actually a need for such a pre (professional exam) test?
I've got it in mind that some (%?) infectious/ill people do not have a raised temperature. Indeed just taking paracetamol can be enough to frustrate such a test!
There are even some who run cool anyway. I am one due to other (rare) conditions so a temperature for me being ill is normal for most!

g3org3y

Original Poster:

22,462 posts

220 months

Thursday 21st May 2020
quotequote all
Benrad said:
How repeatable do you need it to be? What was the difference between the reading with the IR and reading with the in ear?

Maybe try some practise between staff? Take someone's temperature in ear, then take 100 IR readings of the same person by different members of staff and have a look at how repeatable it is. Try different things like whether the patient has sunlight on their face at the time, whether they're sweating or their skin is dry to see how that impacts repeatability
We haven't done extensive back to back testing compared to the ear thermometer. Generally the feeling is that the IR reads low(er).

jet_noise said:
Duh, visor. That's the word for a see through plastic shield of course smile

I would say it is possible to use a thermal camera to remotely monitor from 2m away. Tb would seem to be in the business and be able to confirm. And how much!
An ideal piece of kit would be able to indicate automatically just from someone approaching without human intervention. But that also assumes patients would cooperate!
Is even the stuff used at airports (say) capable of this or is here always human aiming/interpreting?

Lastly is there actually a need for such a pre (professional exam) test?
I've got it in mind that some (%?) infectious/ill people do not have a raised temperature. Indeed just taking paracetamol can be enough to frustrate such a test!
There are even some who run cool anyway. I am one due to other (rare) conditions so a temperature for me being ill is normal for most!
We are probably more proactive than most GP surgeries. We want to do our best to protect our patients and our staff.

Our thought was it's an early screening tool to avoid febrile individuals entering the surgery (and possibly infecting staff and other patients). But for a good screening tool, it has to be accurate, reliable, sensitive and specific.

I do agree one of the major issues (and I mentioned it in the OP) is the asymptomatic Covid patient. No temp doesn't mean no Covid. You could argue that every patient has Covid until proven otherwise and we should take the same precautions regardless.

It's a tricky situation. I don't know what will happen in the upcoming weeks and whether the relaxation of lock down and the public's (understandable) desperation to go out and mingle will result in a second wave. Then we've got the autumn and winter, flu season and all the usual coughs/colds/chesty bugs etc.

Edited by g3org3y on Thursday 21st May 17:48

Benrad

656 posts

178 months

Thursday 21st May 2020
quotequote all
Edit: deleted, repeat post

Benrad

656 posts

178 months

Thursday 21st May 2020
quotequote all
I think you need to 1) prove it does read lower or less repeatably, then 2) either apply an offset or take, say 5 readings and average, or change your technique to improve the measurement

IR could be used to decide whether to do and in ear test or not, doesn't have to be the be all and end all

I'm not an expert, but my day job involves making measurements in a lab (of engine performance/emissions), being confident in the measurement and then deciding how to respond. It's quite a similar situation really and that's how I'd approach your problem, rather than jumping to getting different equipment

Mr Pointy

13,359 posts

188 months

Thursday 21st May 2020
quotequote all
g3org3y said:
Mr Pointy said:
I'm a bit puzzled: nowhere on the page with the specifications for that device does it mention how accurate it is. Surely that's the most critical piece of information you'd look for before buying it?

Given your staff have all the PPE required why not just give them the ear thermometers & tell them to get on with it?
I don't think it's that simple tbh. These aren't clinical staff (HCAs or Nurses) just receptionists. I know other surgeries have their HCAs on the front door doing this.
I don't mean to appear rude but this is a more of a staff management issue than a technology one. If (as you say) you are giving the staff the PPE required to take an ear measurement safely then it's up to the practice management to make it clear to the staff this is how it's going to work. If you get into a situation where the receptionists say "it's not my job" then you'll need to find someone to do it. It's rather disturbing that people working in a GP's surgery think you need to be 2m apart if you're wearing the right PPE.

If you were sending them in unprotected that would be different, but you're not. You already have a reliable method & remote sensing equipment costs thousands so sticking a probe in the ear is the way to go. It will align with the measurements the doctors take as well (within the limits of accuracy of the probes).

g3org3y

Original Poster:

22,462 posts

220 months

Thursday 21st May 2020
quotequote all
Benrad said:
I think you need to 1) prove it does read lower or less repeatably, then 2) either apply an offset or take, say 5 readings and average, or change your technique to improve the measurement

IR could be used to decide whether to do and in ear test or not, doesn't have to be the be all and end all

I'm not an expert, but my day job involves making measurements in a lab (of engine performance/emissions), being confident in the measurement and then deciding how to respond. It's quite a similar situation really and that's how I'd approach your problem, rather than jumping to getting different equipment
If the inaccuracy is constant, I agree it's not not a problem as we could account for it in the measurements.

It appears however that readings can fluctuate notably because of environmental factors. For example, we tested on one receptionist who had come in from the cold and she reportedly measured 33(!). After a few mins inside, she was back up to 'normal' and only 0.2 off the ear reading.

Mr Pointy said:
I don't mean to appear rude but this is a more of a staff management issue than a technology one. If (as you say) you are giving the staff the PPE required to take an ear measurement safely then it's up to the practice management to make it clear to the staff this is how it's going to work. If you get into a situation where the receptionists say "it's not my job" then you'll need to find someone to do it. It's rather disturbing that people working in a GP's surgery think you need to be 2m apart if you're wearing the right PPE.

If you were sending them in unprotected that would be different, but you're not. You already have a reliable method & remote sensing equipment costs thousands so sticking a probe in the ear is the way to go. It will align with the measurements the doctors take as well (within the limits of accuracy of the probes).
I agree. We're quite frustrated about it tbh. We've been very proactive in the kit we've supplied to all our staff, clinical and admin.

At the end of the day, we're not going to push staff (esp non clinical) to do things they're not comfortable with, especially when it comes to risk to health.

Unfortunately Covid has created a lot of angst and worry. When working at the surgery, you're seeing the consequences of Covid on a daily basis. Patients of ours are getting sick and we have had a number of deaths. This daily bombardment is naturally going to cause some level of anxiety. frown