IRSN, Institut de radioprotection et de sûreté nucléaire

Search our site :


Contact us :

En Fr

Enhancing Nuclear Safety


Behind the scenes of dose reduction

Ever better. Optimising paediatric radiology protocols on a daily basis relies on significant work upstream, whether it is a matter of processing data to reflect actual practice or leading epidemiological studies.

Dose reduction not only concerns hospital radiology departments or private radiology practices. Behind the scenes are experts such as those at IRSN. Their efforts serve to optimise protocols, draft standards or even implement tools to reduce the doses dispensed during medical imaging examinations. Demonstration in two examples.

Expertise and national indicators 

Each year, radiology establishments must send a report on the doses received by twenty patients for two routine examinations [1]. Since March 2011 the declarations can be made on-line on the IRSN website, and it is no longer necessary to send them by fax, e-mail or post. The Institute then establishes a national guideline value (mathematically, the 75th percentile, 75% of departments being below this dose): the diagnosis reference levels (DRL).

“This is in no way a regulatory value, but an indicator which reflects practices in France and which enables each establishment to self-assess their system”, insists Patrice Roch, expert in radiation protection in charge of DRL management at the IRSN. “Each country calculates its own DRL, reflecting their own practices.” The Institute has developed the Micado software to help radiologists make their declaration and calculate doses [2].

Paediatric computed tomography requires DRL

The snag is that, even if paediatric DRL exist in conventional radiology since 2004, there is no such thing in paediatric computed tomography, however the examination is known to be more irradiating, because the dose is dispensed over 360 degrees.

The problem is even more complex: “The indications provided by the CT-scanner concerning the dose received by a child are underestimated. The value results from a calculation based on adult bodily volume”, adds Jean-Luc Rehel, expert in medical radiation protection at the Institute. “The smaller the patient volume, the higher the dose dispensed during a CT-scan, with identical settings.” The dose received can be underestimated by a factor of 2.5 for a newborn, according to IRSN's calculations. The solution: industrialists should calibrate their CT-scanners on child phantoms (dummies) and not only on adults.



2011 may be the year of change concerning diagnosis reference levels (DRL). Following joint work with the Institute and the Nuclear Safety Authority (NSA), an order may introduce paediatric DRL in September. Establishments will be held to submit 30 reports for the following weight ranges: 0-10 kg, 10-20 kg, 20-30 kg and adults.

The revision does not stop there. “It defines DRL by scales that are more accessible for radiologists, but the current principle remains that of simple, standard, examinations, limited to a single acquisition, explains Marc Valero, in charge of radiation protection at the ASN. In the future, we should progress towards examinations that resemble radiology practices more, using several acquisitions."

Research and studies

Second area of work: studying the effect of ionizing radiation on children using epidemiological studies. “The studies monitor a population even before a possible disease is declared. This means that enormous cohorts are required for the results to be statistically useful”, says Marie-Odile Bernier, in charge of implementing and supervising these studies on medical exposure at the IRSN.

Typical example: the “Cohorte enfant scanner” (CT-scan child cohort), which initially included 30,000 children from 0 to 5 years having had at least one CT-scan between 2000 and 2006, and which, at European scale (see insert), will include 60,000 additional children from 0 to 10 years having had this type of examination between 2006 and 2013. This will enable the nine participating countries to reach one million children monitored in order to determine if they are at greater risk of developing cancer compared to the general population.


IRSN's contribution to an international project

EPI-CT is a study launched in February 2011 on one million European children. It measures the risk of cancer after a CT-scan.

The project may go beyond the European Union's borders in order to obtain sufficient statistical power for studying types of cancer (brain, thyroid, breast) that are less common than leukaemia. “The IRSN is taking part at different levels”, says Ausra Kesminiene, study coordinator at the International Agency for Research on Cancer (IARC).

“As member of the dosimetry committee, the Institute is working on refining and customizing calculation methodologies: Marie-Odile Bernier is in charge of the descriptive analysis of international cohorts and Dominique Laurier is a member of the coordination committee.”

The Elfe Study launched in April

Another study, called the Elfe study, will monitor 20,000 children born in 2011 in France through to adulthood. The chapter on radiation will study radiological examinations and exposure to radon, in order to evaluate a risk, given this dose. “This study will go beyond averages and will allow further understanding of exposure distribution”, continues Marie-Odile Bernier. In other words, knowing the doses received by the most exposed children, and identifying this specific group of children (socio-economic status of the parents? Regular check ups by a paediatrician or general practitioner? etc.).

Specific studies are limited to smaller populations. One, on radiological exposure of premature babies born in 2002 at Trousseau hospital (Paris), demonstrated that the cumulated dose depends notably on the length of stay (from a few days to several months). It remained however relatively low. Another study, called Carise, on the risk of radio-induced cataract, is for the time being limited to a pilot study on 30 children having had several CT-scans of the skull. However it would be necessary to extend the study to at least 1,500 children to determine the potential effects of radiation on the lens of the eye.


EOS: image and dose optimisation

Because each year, a budget at the Bordeaux teaching hospital (Gironde) is allocated to innovation, the heads of the three radiology departments acquired the first EOS, an X-ray imaging system offering a view of the entire skeleton by 3D simulation.

The advantage: doses divided by 2 to 10 compared to conventional radiology systems.

“We are finalising a comparative study between EOS and a flat sensor [recent system which enabled a significant dose reduction], which I asked IRSN to work on with us”, explains Jean-François Chateil, head of the paediatric imaging department. “The Institute provided us with dosimeters, and offered its expertise in order to determine where they should be placed on patients, and sent us the dose values recorded."

Prior to publication of the results, this healthcare professional revealed that the quality of the EOS images appears to be almost as good as that of the flat sensor, for a lower patient dose, especially in the lowest body areas.


1. In reference to the order of 12 February 2004 on DRL. Examples of benchmark examinations: - radiography: chest, lumbar spine, plain x-ray of the abdomen, mammography; CT - chest, skull, pelvis etc.
2. Micado (dose calculation internet module) evaluates the entrance doses (Ed) that the doctor must declare on the basis of the dose area product or examination settings.



Send Print


Send to a friend

The information you provide in this page are single use only and will not be saved.
* Required fields

Recipient's email:*  

Sign with your name:* 

Type your email address:*   

Add a message :

Do you want to receive a copy of this email?