By Tripp Jones, PhD
I think everyone has seen them by now - the new scanners that the Transport Safety Administration has installed at airports all over the US. As the “radiation person” in my group of friends, I get a lot of people asking about them. “How much radiation do they use?” “Are they safe?” “Should I opt out of the scan and get a pat down?” The answers to the second two questions are subjective, but several studies have been done to evaluate #1. The first studies, done by the TSA itself, met with a lot of criticism from various groups. The study made the assumption that the radiation distributed itself uniformly across the body, which turns out not to be the case. Several members of the UCSF faculty wrote a letter of concern (pdf link), asserting that the risk was underestimated because most of the radiation from low-energy x-rays is deposited in the skin. And since there are nearly 1 billion airline passengers in the US each year, any small increase in cancer risk could add up to significant harm to the overall population
First, we need to make a distinction. There are two types of these new scanners: “millimeter wave” and “backscatter.” The two airports I use the most, Atlanta and Denver, both have millimeter wave scanners. These use the same radiation that your cell phone does, which don’t have enough energy to break chemical bonds in your cells. The backscatter scanner uses so-called “ionizing radiation”, which is the same x-rays used for dental scans or CT imaging. These photons have enough energy to break chemical bonds, and can cause DNA mutations that lead to cancer (more info in Sarah’s post about EM spectrum). From here on I use “radiation” to mean “ionizing radiation” for the sake of brevity.
The radiation dose delivered during one of these backscatter scans has been evaluated in several venues, but the most recent is a paper by Hoppe and Schmidt out of Marquette University. Although the abstract reads like an episode of Ghost Hunters (“Voxelized phantoms of male and female adults and children were used with the GEANT4 toolkit to simulate a backscatter security scan.”), in my opinion these types of papers tend to be very dry (a voxelized phantom is just a computer model of an object receiving radiation). However, the conclusions can be profound – “For a full screen, all phantoms’ total effective doses were below the established 0.25 μSv standard, with an estimated maximum total effective dose of 0.07 μSv for full screen of a male child.” (!)
Okay, maybe that is just exciting to me. The authors did a full Monte Carlo simulation of the radiation dose, which considers the radiation deposited in each part of the body individually (addressing the concerns of the UCSF letter). The maximum dose delivered by the backscatter scanner is 0.07 μSv (less than the 0.25 μSv FDA limit, but more than the original TSA estimate), which raises several important questions, namely: what is a μSv, and how much is 0.07 of them? What radiation is and what it does is another whole topic that could receive its own post, so this is an unfairly brief treatment of it. The Sievert (Sv) is a measure of the “dose” of radioactivity, in much the same way that the amount of a drug (100 mg of Ibuprofen) is the dose. It is based primarily on the amount of radiation energy deposited within biological tissue - the more energy deposited by radiation, the more damage is done to DNA. The Sievert also takes into account the relative sensitivity of different tissues in the body to radiation, as well as the relative damage caused by the type of radiation being considered. So, in essence, the Sv is a measure of how much damage a given amount of radiation will cause. A micro-Sievert (μSv) is just 1 / 1,000,000 Sv.
What many people don’t realize is that we are constantly exposed to radiation in our daily lives. The solar wind generates radiation in the upper atmosphere, which streams down to us on the surface (cosmic background radiation). Traces of uranium in the rocks and soil decay to form radon gas, which seeps out of the ground and is breathed into our lungs. And potassium, one of the building blocks of our cells, is composed of a small amount of radioactive potassium, a remnant of the supernova explosions that generated the matter that makes up our solar system. Together, this is the so-called “background radiation,” which is the amount of radiation we absorb every year, and largely cannot be avoided. On average, the amount of background radiation that people receive is around 3 mSv, where 1 mSV = 1 / 1,000 Sv.
Let’s say you find all this radiation in the environment really disconcerting, and so you build an underground bunker completely shielded from all external sources. You still have some potassium inside you (since you need it to move your muscles), so you can’t fully get rid of all that radiation. This is the radiation you can never escape – the amount you absorb just from living. A pair of French scientists have coined a term for this: the DARI (Dose Annuelle due aux Radiations Internes, or “Annual Dose due to Internal Radiation”), equal to 200 μSv. If you left your bunker, and returned to a normal environment, you would absorb about 15 DARI per year.
As most people know, radiation increases your risk of developing cancer, and we can quantify this risk by studying large populations that have been exposed (for instance, the inhabitants of Hiroshima and Nagasaki who survived the nuclear explosions). From these studies, we have seen that 1 Sievert of radiation increases your risk of developing a fatal cancer by roughly 5%. The data also seems to suggest that this risk depends linearly on the dose delivered, with no threshold. In other words, 100 mSv increases your risk by 0.5%, 1 mSv increases it by 0.005%, etc. Based on this, one DARI would cause cancer in about 1 out of every 100,000 people (compared to the roughly 40,000 people who would have developed cancer normally). This effect of “low-dose” radiation can’t be tested experimentally, but we use this as a conservative, worst-case estimate for radiation protection purposes.
One way to understand the meaning of the 0.07 μSv dose from the TSA scanners is to compare it to other exposures we receive. If we express it in terms of DARI, then one scan delivers 3.5 / 10,000 DARI. I said earlier that we are composed of a small amount of radioactive potassium; if the emitted gamma rays aren’t absorbed by the body, they can actually escape. When I taught Radiation Detection labs, it was fun to show people how they could affect the measurement just by standing too close to the experiment, because their potassium gamma rays were detected by our instruments. It turns out that if you sleep in the same bed as someone for a whole year, you absorb some of this emitted radiation. On average, this adds about 0.00025 DARI to your yearly dose.
You may also recall that bananas are very high in potassium. Since a small fraction of all potassium is radioactive, you receive some exposure from eating bananas. On average, eating one banana gives you 0.0005 DARI of radiation. Some people may also live near a nuclear power plant. Living within 50 miles of one will expose you to, on average, 0.00045 DARI per year. Likewise, coal dust contains traces of uranium; living within 50 miles of a coal power plant gives you about 0.0015 DARI per year. Remember the cosmic background radiation I mentioned earlier? When you fly in an airplane, there is less atmosphere above you to shield these particles. One flight from LA to New York gives you about 0.2 DARI of radiation. Using epidemiological techniques, the lowest radiation dose that has been directly/conclusively linked to an increase in cancer is 100 mSv, or 2000 DARI.
In this figure, I’ve represented various radiation quantities as boxes, where the volume of the box corresponds to the dose. This figure is to scale, so for instance you can fit about 2,000 of the “banana” boxes into the 1 DARI box, because 0.2 mSv is 2,000 larger than 0.1 μSv.
This figure was generated in the same way, only showing the relative differences between different medical imaging scans (and the TSA scan, which isn’t a medical scan). A CT (or CAT) scan involves taking hundreds of individual x-rays in order to reconstruct a 3D image, so these use a lot of radiation. PET scans and technetium stress tests involve injecting a radioactive tracer into the patient, so these also impart a fairly large dose. Hopefully this serves to put the radiation used by the scanners in context. For more information about everyday radiation exposure, you can read the excellent radiation chart made by Randall Munroe of XKCD.
In my opinion, the dose from the TSA backscatter scanners is so low (on the individual level) that it can safely be neglected. The individual cancer risk from this amount of radiation pales in comparison to lifestyle risk factors for cancer like smoking, diet, and fitness. However, one of the central tenets of radiation protection is a concept called ALARA – As Low As Reasonably Achievable. The idea is that one should only use as much radiation as is needed, and no more. Given that there is a perfectly good alternative that doesn’t use ionizing radiation (microwave-based scanner), in my opinion it is irresponsible to use radiation when it isn’t needed. If all 1 billion people who traveled by air in the US per year were scanned using the backscatter device, the worst-case, back-of-the-envelope estimate is that around 4 fatal cancers would be induced (on top of the roughly 500 cancers that would come from the increased radiation during the flight itself). On top of that, the workers who operate the devices are not classified as radiation workers and do not wear any sort of radiation measuring device. Despite these issues with the system as a whole, the radiation from each scan is small, and you shouldn’t have to worry too much about your individual radiation dose when traveling.
Michael E. Hoppe and Taly Gilat Schmidt (2012). Estimation of organ and effective dose due to Compton backscatter security scans Medical Physics DOI: 10.1118/1.4718680