TSA compton backscattering exposes body to non-negligible radiation (2010) | Hacker News Reader

TSA compton backscattering exposes body to non-negligible radiation (2010)

public.asu.edu

178 pointshovden90 comments

This article is from 2010. These days the TSA is using millimeter wave scanners (with frequencies around 30 GHz).

Backscatter x-ray machines are still used in jails and prisons around the US.

In fact, the ones in jails have you stand on a 5-sec slow-moving conveyor belt as you pass through the x-rays, giving thousands of innocent people awaiting bail (among others) a much greater dose than the airports were giving.

There is no opt-out for these people.

Yea I was thinking that. The back-scatter xrays are all gone, but they were not as safe as this article is trying to make. There were a lot of radiologists who were concerned.

The millimeter wave machines are less dangerous, but they also don't really work well at all and have a high false positive rate. They're pretty much security theater, which we're seeing a lot of again now.

> There were a lot of radiologists who were concerned.

Would those be the same group of people who are "concerned" about 5G towers today? You can always find a small group of people who are "concerned" about anything you like. The press uses this for nearly every such story they publish. Without statistics quantifying that "a lot" the statement is meaningless. And even if you have the statistics, there being "a lot" of them is not a guarantee that they were right. One would need to present evidence for their concern as well to be treated seriously. Such evidence does not exist, since airplane flights themselves are a source of radiation exposure.

https://cdn.vox-cdn.com/thumbor/DZqWhneDEMxElDzz7ZiUk0Jv93w=...

A very reasonable heuristic for life is: you should not be concerned about non-ionizing radiation (like 5G), unless in doses that could burn your face off. You should, by default, be concerned about ionizing radiation (like x-rays) in any dose above the natural background.

That's been the heuristic since the 1950s. Now we have much more data on low-dose ionizing radiation. Regulations still assume linear no-threshold, yet no one has definitively shown that anything below 100 mSv causes harm or good using vast amounts of data (c.f. US natural background of 3 mSv/yr). Thus, my rule of thumb is to start worrying only if I get more than 100 mSv either acutely or integrated over a year.

I worry a lot more about eating hamburgers than anything less than 100 mSv.

I agree, and didn't mean to imply otherwise. I was focusing more on the point about ionizing vs. non-ionizing radiation, which I swear to god had to drill into random people's heads in every imaginable way and combination.

I assume this chart is still good to show to people: https://xkcd.com/radiation/?

Gotcha. Yeah lots of people don't get that.

Yeah I still love this chart!

No. People who are concerned about 5G deployment for non-conspiracy theory or ignorance reasons are concerned about unknown effects of increased exposure to nonionizing radiation.

X-rays are very much different from the sub-100 GHz bands used for 5G, and the concerns raised at the time about backscatter xray scanners were that the safety of the systems had not been demonstrated sufficiently to start irradiating everyone who got on a plane, because the negative health effects of ionizing radiation are well understood.

I'm concerned about mm wave 5G because of sideband interference with water absorbtion lines making weather satellites less accurate.

There is the other side, though. You can always find a rather like group of people who will rail against all of the concern and no matter the data will continue to be unconcerned. They often say they advocate for science, but it is completely unscientific to not reexamine one's ideas given new data

I (unlike the OP) did present evidence, though.

X-rays are reasonable to worry about, the radio waves used in 5g, not so much

I haven't commented in a year but I reset my password and logged in specifically to share this: https://www.technologyreview.com/2009/10/30/208491/how-terah... https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4340277/ now, I will again mothball this account

Not before millions were exposed.

Also we're not flying anymore.

With radiation, I always recommend people study page 6 of this DOE radiation dose range pdf [1] to put dose rates in context. The spectrum of radiation doses spans many orders of magnitude and a single atom decaying is readily detectable.

Note in particular that there are places on Earth like Ramsar, Iran that have natural background dose rates above 50 mSv/year and no one has been able to show definitively that dose rates at or below that level do or do not cause harm (or benefit!).

[1] https://www.energy.gov/sites/prod/files/2018/01/f46/doe-ioni...

Another fun data point is the longitudinal studies of flight crew who spend a good part of their career exposed to higher-than-normal natural cosmic radiation (while flying they are above much of the Earth's protective atmosphere) [2]. We should compare how much dose we get from the scanners at airports to how much we get from intergalactic protons raining down from the cosmos while we're up in the air.

[2] https://www.bmj.com/content/325/7364/567

One time I took a geiger counter on a flight just for fun [3]. It really was clickin'!

[3] https://whatisnuclear.com/blog/2014-05-17-radiation-on-fligh...

I don't get why a full year's worth of background radiation is always compared to an acute 5-sec (or less) exposure from an x-ray machine.

If the background rate is 50msv/year, and a machine doses that much (50msv) in 5 seconds, that's 6.3 MILLION TIMES greater than the background dose.

> I don't get why a full year's worth of background radiation is always compared to an acute 5-sec (or less) exposure from an x-ray machine.

Usually because we don't have as much data on acute exposures. The best are from Chernobyl and the atomic bombs (tests and weapons). The issue with these is that we're talking about dosages that are MUCH higher. You'll also find plenty of studies with pilots (and even mechanics), medical staff, and dentists but if you read into the reports you'll see that there are a lot of factors involved that make it difficult to call radiation the culprit, specifically because these groups are exposed to a lot of other known carcinogens and are likely to be tested more frequently (pilots get frequent medicals and doctors are.... well... you know... doctors). So pretty much no matter what you do, there aren't great comparisons and noway to completely isolate just radiation.

Edit: it is hard to get this data because it is unethical to just expose people to radiation and see at what levels they die at. Plus, such a study would take decades

This is an excellent point, does an acute dose of radiation in a short timespan cause more cellular damage than long-term exposure?

Absolutely yes. But acute low-dose below 100,000 μSv has never been shown to cause or not cause harm.

Using uniform units, 0.25 μSv acute dose is what the manufacturers claim for these devices. Annual average background from all sources in US (including these) 6,000 μSv/yr. So you need a scan every 21 minutes for a year to roughly double your annual dose. In some parts of the world, the natural background rate is higher. Like 50,000 μSv/yr. Cancers or other harm have never been shown to occur (or not occur) below acute or annual doses of about 100,000 μSv.

So you need 400,000 of these scans (at the manufactuer-claimed dose of μSv) to reach a dose that is thought to have an above-zero increase cancer risk.

You're not grasping the biology. Damage repair mechanisms have a capacity limit that itself varies with exposure. In particular it's been shown that acute dosage timing of multiple exposures is different than continuous exposure with the same integrated dose.

I will strengthen my statement. Acute doses below 100 mSv have never been conclusively shown to cause or not cause harm or benefit. Certainly annual doses at that level also have not.

The LD50/30 acute dose of radiation is 4,000 mSv. See chart I linked above.

Certainly acute high doses are worse than the same dose spread over years or decades. I learned that in nuclear engineering school 20 years ago. Please don't assume I'm not grasping it.

For a longer term study of the measured cosmic radiation exposure rates that one could expect from commercial flights I have followed this study:

https://spaceweather.com/

They also regularly launch balloons to track atmospheric changes. This is a long-running experiment with data gathered by high school kids. Pretty awesome. You can sponsor a flight and get a token item that has traveled to the edge of space in exchange. Great website.

To estimate the exposure of real flights there is also the "EPCARD ONLINE Flight Dosimetry for Aviation Personel" online calculator[1]. It is provided by a member of the Helmholtz Association, which is the largest scientific organization in Germany, so it should be credible.

[1] https://www1.helmholtz-muenchen.de/epcard/online/fluginput.p...

This is another good chart:

https://xkcd.com/radiation/

This is the chart I always use when my friends start getting uppity about radiation.

It is being proven that 100msv radiation was directly linked with cancer comes from data from Chernobyl, so 50msv I'd still being concerned this is like having 500k X-ray scan at one take, the greter rist doesn't come from the high dose itself, if comes from a time factor, the question needs to be, in how much time you get 50msv? It you tell me in one day I'll very concerned, this is a very high dose for rapidly changing organs like the bone marrow to absorb in one take (causing caners like leukemia ...)

https://xkcd.com/radiation/

Not sure what you're referring to from Chernobyl. The UN team of scientists on radiation (akin to the IPCC for climate change) has this to say about Chernoby: https://www.unscear.org/unscear/en/chernobyl.html

The open letter from biophysicists at UCSF _at the time_ outlined many of these concerns [1]. It was clear that the way these machines were being assessed was not scientifically rigorous, and were likely dangerous. The DHS Secretary Michael Chertoff had an interest in the company selling the scanners [2]. The damage done to the public’s trust and the public’s DNA by forcing these machines on people was awful.

[1] https://www.thestranger.com/images/blogimages/2010/11/17/129...

[2] https://en.m.wikipedia.org/wiki/Michael_Chertoff

DHS is a far bigger threat to our freedom than terrorism.

In "policing" work, there is a line where you become a bigger threat than the bad guys, and you become the bad guys.

Many DHS agencies don't care. They are self-perpetuating institutions that act in their own "institutional" interest.

Note to DHS language parser: this comment should be classified into your "big" database ("normal patriotic American complaining about gub'ment"), not your "small" one ("actual threats").

Has anyone seen these recently? I saw one in Boston many years ago, but every airport I've been through since uses the mm-wave ones, distinguishable by their rotating action

They were retired before May 2013.

Officially retired due to the "FAA Modernization and Reform Act of 2012" which banned scanners that produced nude images, but the scanners came under continuous criticism for safety and were banned in the EU and elsewhere.

I'd be surprised to see them try and re-introduce backscatter x-ray scanners when the less divisive Millimeter Wave scanners work "well enough" for their security function. Backscatter was just a political lightning rod.

The Wikipedia article goes back and forth on the safety aspect:

https://en.wikipedia.org/wiki/Full_body_scanner#Backscatter_...

It boggles my mind that people accept to be treated as lab rats with unknown consequences to their health decades down the line, just so an utterly useless bureaucracy can justify its existence and certain political goals can be met.

Of course the TSA is just the tip of the spear. As James Duane so powerfully demonstrates in his talk and book [1], the number of gun&badge toting Federal agencies with the power to ruin one's life, sometimes even making up crimes to do so, has been steadily increasing. As has the number of nonsensical criminal statutes and regulations in the books.

[1] https://www.cato.org/events/you-have-right-remain-innocent

Since their introduction, I've had no issues telling the agent "I'd like to opt out of the scanner". This adds ~10 minutes as I have to wait and perform a manual pat-down, but personally that's not something I mind. Others obviously may feel otherwise for varying reasons.

I've found this article (10 years old now, : https://www.health.harvard.edu/diseases-and-conditions/are-f... ) to be fairly well-rounded. I wonder if, since then, there has been a better analysis of actual/measured results of these technologies.

I have always opted out of these scanners as a matter of principle. (To wit, I think it’s an erosion of privacy.) But you should know that the TSA changed their guidelines such that granting a request to opt out is effectively up to the discretion of the TSA agent at security. Their new guidance says:

>While passengers may generally decline AIT screening in favor of physical screening, TSA may direct mandatory AIT screening for some passengers.

I found out about the change as I was going through security and opted out. The folks in line behind me said “I didn’t know you could do that” and a queue formed of folks opting out. They made everyone else behind me go through the scanner, saying that it was a security risk to have folks queueing to opt out. (It’s a security risk to form a queue inside a long security screening queue?) The reality is that this rule change was needed so that they could avoid the screening process grinding to a halt if everyone requests a manual screening.

https://onemileatatime.com/tsa-scanner-opt-out/

Interesting. In EPIC v. DHS, the US Court of Appeals for the DC Circuit ruled that the TSA body scanners were constitutional, partly because "any passenger may opt-out of AIT screening in favor of a pat-down, which allows him to decide which of the two options for detecting a concealed, nonmetallic weapon or explosive is least invasive."

It seems like anyone who is denied the right to opt out could have a good case against the TSA in court. This Forbes article says: "Contrary to TSA’s statement in its new policy, TSA may not legally direct mandatory AIT screening for any passengers. Not according to EPIC, and not according to TSA’s own lawyers. EPIC v. DHS requires as a condition of a finding of constitutionality of the TSA screening procedures the ability of passengers to opt for pat-downs in lieu of AIT screening."

https://www.forbes.com/sites/lisabrownlee/2015/12/24/tsa-bod...

The nudie-scanners have been gone for over six years and the current scanners are nothing more than (at a high level) low-resolution "difference" detectors.

The only way they'll invade your privacy is if your genitals are made out of metal. And no, genital piercings don't trigger them.

With ATR (automated target recognition) scanners it is literally, definitely, assuredly, irrefutably impossible to reconstruct any image or representation of the human body in any capacity approaching useful or attributable using 2020 scanners.

What are the privacy ramifications of 2020 airport scanners?

Not 2014 scanners. I do not care about something that does not used anymore.

I would assert that a patdown is an infinitely greater "erosion" of privacy than the post-feature and contour extraction comparison of phase history data to a generalized model of the human form.

Who cares if there aren't privacy implications? It's still security theater and should be opposed on principle. Anything beyond metal detectors is pointless. You can't hijack airplanes anymore because the other passengers won't comply with hijackers anymore; that was proven on 9/11, long before any fearmongering bureaucrats imposed any of their corrupt theatrical horseshit.

In what way is it security theater?

It’s real and it works better than the technology it replaced.

Would you be interested in purchasing a tiger-repellent rock from me? I can guarantee that if you purchase one of my rocks, no tigers will eat you. They're real rocks, I assure you.

I tried to refuse walking through yet another xray upon arriving in Laos in 2014 and 10 agents appeared out of nowhere. So you walk through the xray.

None of this shit is even necessary in the first place. In more civilized countries you can be through the whole security screening before you even realize it. Pivot the make-work program to casting and smashing bricks, instead of inconveniencing everyone else. They could even keep their fancy uniforms.

Do you also opt out of millimeter wave scanners? I did opt out of the backscatter ones, but do not do that anymore.

I do, largely for political reasons, partly for trust reasons.

Or just get pre check and 90+% of the time you’ll be quickly shuffled through a metal detector. Then opt out the rest.

yes, opt-out other than for a metal detector.

but no, shun pre-check, another form of surveillance-conditioning.

Pre-check isn't surveillance conditioning, it's just a cynical cash grab, where you pay to solve a problem that the person you are paying is forcing onto you.

It's included with Global Entry which is a HUGE time saver. Don't hate on it just because you have to pay. I have my disagreements with TSA like everyone but my time is valuable and if it costs 110ish bucks for four years that's more than worth it to me.

that's true too, but it's not either-or, it's both.

Can you expand on that? I got precheck and was pretty happy with the process. They genuinely seemed to be making sure i wasn’t up to something strange before granting me the status.

exactly. it conditioned you to accept that submitting to and answering a bunch of invasive questions is totally acceptable, even happiness-inducing, in exchange for "safety", where safety is actually systemically guaranteed simply because very, very few are malevolent enough to attempt mass murder, or even just one murder. it plays on mental biases that more easily extrapolate imminent and colossal danger from rare incidents than vice versa.

sure, the invasiveness seems minor on an individual scale, but it's massive on a population one. power is more easily amassed in little morsels across grand scales than taken by force in large chunks.

You give up biometrics and background information. The interview is designed to be intimidating.

That wasn't my experience. I sat in a waiting area for about 5 minutes, then I sat down at a desk with an "interviewer" who pulled up a screen with the information I'd already entered when I applied. She asked me if it was true, I said yes, and then she said there'd be a background check and I'd get the result in a bit. It was less intrusive than a credit application.

During my first trip through $100 precheck (via global entry) I was directed through millimeter wave machine. I opted out. Precheck was not prepared for opt out. In the end, it took longer to get through pre-check than the normal+optout.

Fear is a powerful motivator.

If everyone opted out of the scanners, they would quickly disappear.

I opted out of backscatter back in the day (and looking at the facts retrospectively I feel validated in doing so, for example the miscalibrated units found to be spitting out 20x more radiation than designed).

I don't opt out of Millimeter Wave today, as I'm yet to see convincing scientific arguments or evidence that they're unsafe, or what the mechanism would be outside of mild tissue heating. I do choose to close my eyes when scanned though, take that as you will.

Have you seen any convincing scientific evidence that they are effective? I mean... Do they stop terrorism? They may not give you cancer but those machines aren't free.

Beyond the scope of opt-outs. This is more a political question masquerading as a scientific inquiry.

Yup. I have opted out every time. Maybe 50 times since they came out? Just my 3 seconds of civil disobedience.

counter point: it everyone opted out they would become mandatory

If enough opted out they would need to find another form of security theatre.

that's debatable, but my feeling is it's headed towards mandatory anyway so at least if everyone started opting out we have a chance we can force them to get rid of it.

Unlikely, politicians don't like pissing off everyone.

No, they appoint others to handle that for them. In this case, the TSA.

These scanners did disappear. They aren't the current millimeter wave scanners.

In every airport of every country?

When we sold our last ICE car to go electric, we learned that a lot of used ICE car inventory is being shipped to Mexico and other countries where they are still ok dealing with the smell (or just not in a financial position to move forward as fast). I imagine the same kind of dynamic happens with equipment like this.

I definitely only meant it for the TSA, and I'm glad you pointed this out.

However, I suspect that in these countries that have the financial constraints you describe, opting out may not always lead to decommissioning them because installing alternatives requires significant expenditure (which is why the US could afford to get rid of them in the first place). I don't think this is universally true (i.e. I expect some countries ended up with old backscatter scanners but would choose to replace them if many people opt out) but I don't know that this intersection is very large.

It's not like ICE cars (both new and used) aren't still being sold in tremendous amounts in the US…

Used ICE car prices are crashing though.

there's no value add - other countries aren't really concerned with security theatre to the extent that the US (maybe UK?) is. witness the difference between screening at e.g. international airports outside of US/UK and then additional necessary screening for US bound flights.

perhaps, though, that makes it a great export opportunity there, similar to arms exports: US law will mandate that these scanners are minimally required for inbound flights, other airports pick them up in order not to lose US flight slots.

I think something like 4-5% of car sales in the US are electric-only or hybrid, the remaining 95% being old-fashioned ICE. It's not just Mexico still buying them.

Yeah. It’s still early days.

I'm not sure if they were ever in use outside the US.

I have seen a few at airports outside the US after we switched to millimeter wave. So yes, they are used in other countries. (And unlike the US, opting out sometimes isn’t an option.)

I really don't worry about the low levels of radiation or not.

What I worry about is trusting the TSA to keep these machines operating correctly and within spec. The care they give the machines cannot be greater than the care they give to the employees, and we know how well they treat employees.

This article is a little absurd. They are talking about uSv (10^-6) than give the big warning

> Scary thing, what happens if scan jams and fail-safe mechanism fails -- local dose very high of order a few Sv

We can do this with anything. I'm not so concerned about these machines failing in a region that typically has radiation monitors. Sure, we should be concerned with machinery failing, but the amount one should worry isn't proportional.

It’d be fascinating to know how many products, or entire businesses, are able to exist or even thrive through broad ignorance (whether willful or naïve) of their true nature.

.emf files on a web site?

There's a reason that the scanner manufacturers got exemptions so they wouldn't have to report measured radiation.

Needs to be retitled (2010)

Looking forward to the epistemological study in my 70’s after I’ve had my prostate ripped out, wherein we find men who traveled frequently post 9/11 were at increased prostate cancer risk from millimeter wave tech.

I can’t recall the name of the principle, or the exact dose, but isn’t 1 Sv enough to cause a 50% death probability from radiation sickness, and 3 Sv enough to cause a cancer? There is a good youtube series (IIRC nuclear policy course at MIT) I was watching when the Chernobyl show was on last year.

So 0.25 mSv is enough to cause a cancer for every 12,000 people they scan. That’s a lot of people considering the throughput of airports. Lovely. I’m glad this was shit-canned.

The LD50/30 acute dose of ionizing radiation is around 4000 mSv (4 Sv).

You're assuming a linear no-threshold dose response for ionizing radiation, which is indeed what radiation people have used since the 1950s when we didn't have enough data to be less conservative.

In the US, natural background dose averages around 3 mSv/year. Some places in the world have high backgrounds, in the 30-100 mSv/year range. Even there we have not seen a clear negative (or positive) response.

I get the point about linear no threshold, but I don't get the point of bringing up background dose. If there is a threshold, then acute doses are worse than cumulative right? Taking even your highest figure 100msv, that's over an entire year. In a 5 sec span, that would be 0.0000159 msv, or 200,000x less than the 3msv from the machine in 5 sec (or less).

Acute high-dose is indeed worse than cumulative. Less than 100 mSv acute has never been shown to cause (or not cause) harm or benefit.

I bring up background just to point out that radiation is very natural and our bodies evolved in a radioactive world and have repair mechanisms that appear to be totally capable of handling low doses relevant to these machines. I can see that that's confusing to a degree.

I may be mistaken, I believe you're confusing milli-Sieverts (mSv) with micro-Sieverts (μSv). The machines are spec'd at 0.25 μSv/scan, or 0.00025 mSv.

You have 4 litres of blood in your body.

You will die if you lose 2 of them.

A typical lab blood draw is ~10 ml.

Does this mean that for every two hundred people who get their blood drawn, one is killed?

Of course not, but if I inject you with 2 L of peanut oil, it will kill you. If I do the same at 1 ml to 2000 people, maybe the statistics work out so a single, hyper-allergic person is dead. That’s all the linear approximation is saying, and that lecture series says it has held true. The interesting thing about it, is the dose for a single cancer is above the lethal dosage.

This is a great example of the ridiculousness of linear no threshold models.

Reply on news.ycombinator.com