As I emerged from the aerobridge into the Manila terminal I noticed a woman pointing a weird device at me. It was a thermal camera, a gizmo used to measure skin temperature. High temperature areas show up yellow or red while low temperature areas are blue or purple. Given the current paranoia about things like bird flu and SARS this camera was being used to screen passengers in case someone had a fever, which could be indicative of an infection with one of these diseases. If you read the literature put out by the company that produces it (http://www.flir.com/cs/apac/en/view/?id=42362) they would have you believe that this is an effective screening process. But is it?
Thermal cameras have been used in veterinary medicine to detect lameness and abscesses by the heat they generate. You will no doubt have sprained something at some point and noticed how warm the area became, because of the inflammatory process. This is different to the central fever that is generated by infectious agents. According to the company’s literature the temperature in the corner of the eye most closely approximates core body temperature. The literature also states that “glass and plastic do not transmit infrared radiation, so people need to remove their glasses in order to be examined.” That’s me slipped through the net. The recommendation is also to scan individual passengers at a distance of 1 to 1.6 metres so that the face fills the screen. How this is possible for the mass of humanity streaming off the plane, where some people are looking ahead, others are looking to the side and still others are looking at the ground, is beyond me. However, this is more an operational fault than a fault with the camera per se.
A recent study attempted to objectively evaluate the effectiveness of thermal cameras in detecting fever. In this study all subjects that wore them removed their glasses and stood still in front of the device for two to three seconds, very different to the moving people at the airport. Using this method the thermal camera had a 90% sensitivity (10% of the people who had a fever were not detected) and an 80% specificity (20% of the people who were detected as having a fever did not actually have one) (Nguyen et al 2010). These figures were deemed to be acceptable even though it meant that some people escaped the surveillance net while others would be falsely detained. Still no system is 100% fool-proof.
What also puzzles me is what the authorities would have done if they had registered a positive. Presumably that person would have been asked to stand aside for further investigation, but what of the other passengers? Would all the previous passengers, who had by now dispersed through the terminal, be rounded up? Would all the passengers yet to emerge from the plane be detained? Surely if one person had a fever then all the passengers and crew would need to be regarded as infected, as we had all been breathing the same air for the past however many hours? Veterinary medicine is much more straightforward and logical. Any animals shipped overseas need to have a defined set of vaccinations beforehand (e.g. rabies if coming from a rabies endemic country) and they enter a period of isolation after arrival at their destination. If they are healthy at the end of this quarantine period they are free to go. This is of course completely impractical for people, who like to jet-set around the world on a daily basis. However, it certainly does enhance the potential for diseases to spread rapidly around the world as we saw with SARS, which travelled at the speed of flight from Asia to Canada.
The other problem with thermal imaging is that it only detects infected people who have a fever. Many diseases, such as bird flu, have an incubation period of several days during which the infected person displays no clinical signs but is still potentially infective.
I don’t know what the solution to averting a pandemic is, possibly a combination of individual country surveillance of people and animals coupled with vaccination. Infrared thermography has a potential role to play but only if used properly and not in the haphazard manner I saw in Manila. Unfortunately this would add to passenger inconvenience, which would, however, be a small price to pay if it might avert or mitigate a pandemic.
Dr. F. Bunny
Nguyen, A.V., N.J. Cohen, H. Lipman, C.M. Brown, N.A. Molinari, W.L. Jackson, H. Kirking, P. Szymanowski, T.W. Wilson, B.A. Salhi, R.R. Roberts, D.W. Stryker, and D.B. Fishbein. 2010. Comparison of 3 infrared thermal detection systems and self-report for mass fever screening. Emerging Infectious Diseases 16: 1710-1717.