A new study of airport body scanners by U.S. Army scientists shows that the machines produce a low dose of radiation, supporting Transportation Security Administration claims that a screening is equivalent to the radiation a passenger gets in two minutes of flying.

The scanners, which use X-rays to check for objects hidden under clothing, have been the subject of controversy about how safe they are and whether they create a cancer risk for the traveling public. Although the study is unlikely to douse those concerns, one critic of the machines called it the most reliable test to date.

"These are the best measurements that have been done," said Peter Rez, an Arizona State University physicist who in April signed a letter to the White House science adviser questioning previous tests. "Things have to be filled in, but it's a step in the right direction."

The study results obtained by ProPublica were presented at a Health Physics Society conference in late June.

The TSA rolled out plans to put full-body scanners at nearly every security lane by 2014 in response to the attempted bombing of a Detroit-bound plane by a man who had explosives hidden in his underwear.

The agency uses two kinds of imagers. In the backscatter machine, a passenger stands between two large blue boxes and is scanned with a narrow X-ray beam that rapidly moves left to right and up and down the body. In the millimeter-wave machine, a passenger enters a chamber that looks like a round phone booth and is scanned with radio-frequency waves.

Although the safety of both machines has been questioned, most of the attention has centered on the backscatter machine because it uses ionizing radiation, which can damage cells and cause cancer.

To reassure travelers, the TSA asked the Army Public Health Command to conduct radiation surveys at airports around the country. The new study, paid for by the TSA and done at a TSA lab, was part of that work.

There is much debate in the field about how little—or whether any—radiation is safe.

According to the Army presentation, the average dose to the body was 3.8 microrems per screening, meaning that a person would have to go through the machine more than 5,000 times to exceed the annual dose limit recommended by the American National Standards Institute. By comparison, a chest X-ray produces about 10,000 microrems of radiation.

The test also provides new information about the dose to the skin and eyes, which has been a concern because low energy X-rays deliver a large portion of the radiation to shallow tissues. The Army testers measured the average dose to the lens of the eye at 6.7 microrems and the average skin dose at 11.3 microrems, both of which were extremely low compared to the accepted guidelines.

The study, however, does not make a conclusion about whether the machines are safe.

Some scientists have estimated that the routine use of the machines planned by the TSA could result in anywhere from a handful to a hundred additional cases of cancer over a lifetime. But even with long-term tests, it would be difficult to separate the scanner effects from other common sources of exposure, such as medical X-rays and cosmic radiation from flying at high altitudes.

The test is different from others that have been done because the Army used "optically stimulated luminescent" dosimeters—similar to the badges worn by nuclear workers—to measure the radiation.

In the past, the TSA has relied on tests conducted by the Food and Drug Administration and Johns Hopkins University that used a radiation detector known as an ionization chamber. Rez and a group of scientists at the University of California, San Francisco, argue that ion chambers are inadequate for the fast-moving, low-energy beam emitted by the backscatter machines.

An ion chamber is a device that measures radiation exposure by generating an electrical current. In contrast, an OSL dosimeter is a new type of badge that uses a crystal, which is then read with a light.

Usually, dosimeter badges aren't adequate for a single scan because they need a high dose of radiation or an exposure over a long period of time to register a reading. Walking through a body scanner on the way to catch a plane would probably not produce a reading.

To overcome that problem, the Army used robots to run 93,000 screenings over two weeks. They simulated a 190-pound person by building a dummy out of water jugs arranged on a wooden frame.

Testers then placed 181 dosimeters in and around the system, according to an email ProPublica obtained from the Electronic Privacy Information Center, a civil liberties group in Washington, D.C., that sued the government to get scanner records under the Freedom of Information Act.

Rez said the study doesn't address his concern about how much radiation a person would be exposed to if the machine breaks down and the beam stops on one spot on the body. It also didn't stanch the calls from Rez and others to allow testing by outside scientists.

"From a public relations point of view," Rez said, "it would have more clout if it was done by someone who can be seen as being independent of government."