Checking for Fever: A Comprehensive Guide to Implementing a Temperature Screening System

Answered November 06 2020

maintenance technician performing fever screening

The Coronavirus pandemic of 2020 is not the first time infrared has been used to screen people for fevers, but for some reason, it was the first time people bombarded me with questions about it. Everyone from security guards to grocery store baggers wanted to know more about fever screening. My problem at the time was that I didn’t know more; I’m a predictive maintenance girl, not a doctor. So I set out on a quest to learn, and as usual, I began by digging into history.

A Brief History of Using Infrared for Temperature Screening

As far as I could find, infrared was first used to screen for higher body temperatures in 2003 during the SARS epidemic in China. Official Chinese government data indicates that during two months, 30 million travelers were screened in China. Infrared was next used during the 2009 H1N1 outbreak. This was followed by MERS-CoV, the Middle East Respiratory Syndrome Coronavirus, in 2012. It has been used in Taiwan during outbreaks of Dengue, and it was used in 2014 for the Ebola outbreak that originated in West Africa.

As I write this, in September of 2020, controversy still looms large on social media platforms like LinkedIn regarding the effectiveness of thermography for fever screening. History and empirical data have demonstrated that infrared is useful in screening for fevers when used in compliance with appropriate international standards and guidance from the World Health Organization.

The beauty of using infrared for febrile (or fever) screening is that it takes person-to-person contact out of the picture. With some systems, you can run the screening process from a completely different room, so it increases the distance between the screener and the “screenee.” It is also faster than using a thermometer on every single person who walks into a building.

Guidelines and Standards You Should Know

Before everyone gets excited, put down your Fluke Ti450 or Flir One because there are guidelines that you should know about that most suppliers don’t want to tell you. My go-to for standards is almost always ISO, the International Organization for Standards. For fever screening using infrared, there is an International Standard numbered IEC 80601-2-59. Read through that, and you will learn that the cameras you use for predictive maintenance are probably not going to cut it. What is even more alarming? Most systems sold as febrile screening equipment do not meet that standard either.

Most cameras used for CBM (condition-based monitoring) or PdM (predictive maintenance) activities have an accuracy of +/- 2⁰ C. For fever screening, the camera should have a measurement of uncertainty not greater than 0.5⁰ C over the range of 34⁰ to 39⁰ C. I hope that paused you a bit. When I first started digging into fever screening with infrared, I honestly thought that such a camera didn’t exist.

The Difference Between Core and Surface Temperatures

One of the most important things to remember on this topic is that an IR camera measures surface temperatures and not core body temperatures. Do you remember your parental figure sticking a thermometer under your tongue when you were a kid and telling you to be still and not bite the thermometer? Well, the under-the-tongue or…eek…the rectal thermometers are measuring your core body temperature. Core means the temperature inside of your body. External surface and internal core temperatures are not the same.

Infrared can only see surfaces. It can see human skin quite well, but it cannot see through human skin and fat to determine core internal temperatures. So, at room temperature, the surface temperature of the human skin is lower than the inner core temperature, even when a fever is present. So, forget everything you think you know about checking for fevers and remember that to find a fever with infrared at room temperatures where the subject has had time to adjust to ambient temperatures, you’re looking for temperatures above 35⁰C or 95⁰F rather than the old fashioned thermometer temps of 100⁰ F and above.

The surface temperature of human skin is also significantly impacted by environmental and ambient conditions. Imagine you were walking around a parking lot in Phoenix, Arizona, in August. If someone took an infrared image of you in the parking lot, you would probably look feverish. Not only that, but high temperatures also tend to make people sweaty. Excess sweat on the skin mucks up emissivity.

Now consider the other end of the temperature scale and imagine you are in Caribou, Maine, in the middle of January, and you have the flu, but you had to run to the drugstore for some medication. Suppose the drugstore was conducting fever screenings. Immediately after entry, you would appear so cold that an illness wouldn’t be readily recognizable.

Fever screening works in places like airports because your body has time to adjust to the ambient temperature inside the airport. It takes an absolute minimum of five minutes for the human skin to adapt to indoor ambient temperatures. Ten minutes would be ideal. Letting folks hang around for five to ten minutes before screening can be challenging to achieve, depending on the layout of the building and where you want the screening to take place.

Guidelines for Effective Febrile Screening

Febrile must be performed indoors. It doesn’t work outside because you can have varying atmospheric temperatures and humidity levels. If you were thinking you could do a screening outside your building before allowing people to enter, it’s time for some rethinking. Anyone who has performed electrical inspections outside can tell you that when you conduct infrared inspections outside, you must deal with wind. The wind is the bane of all outdoor inspections, wreaking havoc on the accuracy of surface temperature measurements. Indoor temperatures are a lot more stable and tend to be carefully maintained.

To conduct viable screenings, the ambient temperatures where the screening is conducted should ideally be room temperature. Room temperature is between 18 and 24 degrees Celsius, or 64.4 and 75.2 degrees Fahrenheit. Don’t forget to consider humidity. The ambient humidity needs to be between 10 and 75%. Ideally, it would be slightly below 50%.

The thermal image needs to be of the face. I’ve seen a lot of posts on LinkedIn where the thermographer has taken a picture of the ear, or sometimes they post the forehead’s temperature. ISO recommends the image be taken at the inner canthus of the eye. The inner canthus is the part of the eye closest to the nose near the tear ducts. Aim the camera for the corner of the eye closest to the nose, and you’ll be at the inner canthus. This area will provide the most accurate and consistent surface temperatures on the human body. This accuracy is due to the slight cavity effect of the tear duct and the proximity of the internal carotid artery.

To accurately image the inner canthus, all obstructions and shadow casters have to be removed. That means people being screened will need to remove their glasses, eyewear, hats, and any large hair adornments that cast a shadow on the face. Some people dislike doing that, which may be why I see so many images of ears floating around on social media.

There are some other things you need. A blackbody is required to confirm the accuracy of the camera. Ideally, it should be in the image with the person being screened. The other option is to take a thermogram of the blackbody immediately before the thermogram of the person being screened.

Tips for Selecting Equipment

Then there’s the question of the camera. We already know it has to be extremely accurate, but there are some other must-haves:

  • The camera must have a target plane equal to or greater than 320 by 240. Ideally, the face should fill at least 240 by 180 of those image pixels. 
  • You’ll want a camera with a non-linear pallet, like the rainbow pallet. 
  • The displayed temperature increment can’t exceed 0.1⁰ C. 
  • The date and time have to be recorded either on the image or in the image’s properties. 
  • The camera has to have a non-uniformity compensation (NUC) device. 
  • The camera must also allow the user to set up a threshold temperature and a warning temperature. (Hello Isotherm, my old friend.)

I recommend the camera and blackbody be calibrated at least annually by a 17025 certified lab. That means you may need to have two of each if you have a single entry point to your facility. Some systems only require the blackbody to be calibrated and not the camera, so check the specs.

Things to Watch Out for with Suppliers

I have had several suppliers tell me a blackbody and my other target specs were not necessary. Those suppliers will usually wave around the flag of ISO 80601-2-56 or 21 CFR 880.2910, both of which are standards for clinical thermometers. Infrared thermography does help us visualize temperature, but it is not what I would call a thermometer.

In one frustrating case, I asked a supplier for specifications, and instead of sending me what I requested, I received a reply telling me I didn’t need the specifications because their system had a Premarket Approval (PMA) from the Food and Drug Administration (FDA.) It took me a lot of effort to acquire that system’s specifications, and in the end, I was unsurprisingly disappointed.

If you are considering using infrared for fever screening, spend the 280 bucks and download the ISO IEC 80601-2-59. Don’t let the madness on social media lead you astray, and don’t let some pushy salesperson sell you something that doesn’t fit what you want. Keep in mind that a salesperson’s job is to sell things. I realize that I’m a bit of a purist when it comes to both thermography and standards, and I understand you might not have the cash flow or the space available to comply with 2-59.

If you are in that boat, please proceed with caution. There are some rather nefarious folks out there trying to make a buck from the pandemic panic. Choose a standard and specifications that work the best for your needs and stick to it, don’t let the supplier blindside you with obscurity or convince you that you don’t need what you want.

Suggestions for Implementing a Fever Screening System

If you have someone in your organization with infrared experience, even if it is PdM in nature, pull them into the decision-making and installation process. If you don’t have such a person, then educate yourself on some infrared basics and be sure you know your standard and what specs you want for your system.

When you choose a system, and it finally arrives at your facility, follow the manufacturers’ set-up and installation instructions to the letter. Try to install it in a location that will allow people time to adjust to the temperature in the building. That last bit can be very challenging, depending on the layout of your entrance. Mark the floor for the screening area, so folks know where to go. You might also consider some signage that briefly explains the process and ensures the individuals being screened that the process is safe. If you are screening your employees, send an email in advance to let folks know what’s going on. If you can, put a second display screen in the screening area, so people can see themselves in infrared.

Create a process for what happens if the system flags an individual as having a fever. Have an isolated, climate-controlled location where they can wait five to ten minutes. Perhaps they were in that Arizona parking lot and needed a chance to cool off a bit. From that point, you can try another infrared scan or jump straight to a secondary, more traditional temperature measurement. If they pass, it’s business as usual. If they have a fever, though, have a plan and procedure for what comes next–both for them and for any other folks that they were near.

Conclusion

Fever screening with infrared thermography isn’t new, and it’s an effective method of fever screening. With so many system options available, be sure to pick a standard and stick to it. Remember that surface and core temperatures are not the same, and that the environment impacts surface temperatures. When screening, target the inner canthus of the eye. Get help if you need it or get educated if you can’t find help, and finally have a procedure for the screening process. I hope that helps, and I wish you the best of luck in your efforts if you choose the path of using infrared for fever screening.

The views expressed in this article are my own and don’t reflect the views of my employer. 

_____________________________________
Vestanna McGuigan headshot

Vestanna McGuigan
Vestanna is a Reliability Engineer with a strong passion for infrared thermography. She is skilled in inventory control, PM optimization, and failure mode effects analysis. Vestanna is also knowledgeable in predictive maintenance tools and is currently developing maintenance strategies and processes for reducing maintenance losses using proven technical tools. She currently lives in Topeka, Kansas.

Asset Management Questions & Answers