Informatics Tools for Radiation Dose Estimation

From Clinfowiki
Jump to: navigation, search

Informatics Tools for Radiation Dose Estimation


The proliferation of medial imaging over the past several decades has sparked increased interest from health care providers and the general public regarding patient radiation exposure and the associated long term radiation-related health risks (1,2,4). The most feared long-term complication of medial radiation exposure is the development of radiation-induced cancer. Ionizing radiation can damage DNA or produce free radicals that result in DNA injury. Cancer can result when injury to the genome is not adequately repaired by the bodies natural DNA repair mechanisms. The development of cancer related to radiation typically occurs one-two decades following exposure (4). The main culprits in medical imaging resulting in the greatest amount of patient radiation exposure include computed tomography (CT) scans and examinations utilizing radioactive isotopes (nuclear medicine)(2,6).

Radiation Dose

Radiation dose can be expressed using numerous metrics. In medical imaging, doses are most commonly expressed in millisieverts (mSv). This unit of measurement takes into account not only the amount of radiation delivered but the relative biological effect of the subtype of radiation (4,8). Available population data regarding radiation exposure and the associated risk of developing cancer demonstrates a clear causative relationship at exposure levels above 100 mSv (5). More specifically, the literature has shown about 5% excess risk of death from cancer with a dose of 1000 mSv. The risk of radiation-induced cancer is controversial at levels between 10 mSv and 100 mSv and even more questionable at exposure levels below 10 mSv. Regardless, experts believe that the risk of radiation-induced cancer follows a stochastic model, meaning that it can occur at any level and increases in likelihood with increasing dose (8). As mentioned above, CT examinations (exposure ranging from 2.0 to 10.0 mSv) and nuclear medicine studies (exposure ranging from 2.0 to 12 mSv) currently contribute the most to patient radiation exposure. As a reference, a single chest radiograph (x-ray) has a radiation dose of 0.1 mSv and the yearly background radiation dose an individual receieves from the environment is approximately 3 mSv (2,4,9).

Informatics Tools

Informatics applications developed to capture and display patient radiation exposure can have a tremendous impact on decreasing the inappropriate utilization of ionizing radiation in patient care. New legislation aimed at reducing radiation related complications now mandates that radiation dose information be documented in the Radiology report. Effectively extracting this dose information and communicating this information to interested parties has been historically difficult. Frequently, both the dose information transmitted from the imaging modality (CT scanner) and the dose information found in the imaging report are stored in not easily queryable image or text-based formats. Informatics tools utilizing optical character recognition (OCR) have been created to transform dose information from image or text-based formats into more usable data until dose information is more consistently incoporated into the DICOM message format.

Researches at the Brigham and Women's Hospital have developed an open-source application refereed to as GROK (Generalized Radiation Observation Kit) which performs just this function for CT examinations. This tool is not only adept at capturing current and future radiation exposure, but is able to retroactively query the electronic health record (EHR) for a patient's prior examinations to calculate cumulative radiation dose estimates (3). Once radiation dose information is available in a digitally usable manner, it can be used to power decision support systems in the EHR and to ensure that the institution is meeting governmental regulatory benchmarks. Facilities can then upload the digital dose information to the American College of Radiology (ACR) Dose Index Registry which will then provide them with periodic analysis of exposure rates compared to other regional and national values (3).

Health system wide and even national or governmental informatics tools used to capture radiation dose information can ensure accurate cumulative radiation dose estimates as patient's are imaged over time at different hospitals in different parts of the county. Ultimately, radiation dose estimates and the associated projected risks are less usefull unless they account for all of a particular patient's examinations performed with ionizing radiation (3).


In general, the benefit of performing a radiological examination with ionizing radiation outweighs the associated risks. This principle has recently been questioned in the light of potential overuse of medial imaging in patient care. Informatics tools can aid the clinician in making wise imaging choices that maximize the available information which minimizing the radiation dose received by the patient.


1. Amis, ES, Butler, PF, Applegate, KE et al. American College of Radiology white paper on radiation dose in medicine. J Am Coll Radiol. 2007; 4: 272–284

2. Brenner, DJ and Hall, EJ. Computed tomography:an increasing source of radiation exposure. N Engl J Med. 2007; 357: 2277–2284

3. Jonathan S. Batchelor. Systems Mine Patient Records to Capture CT and Nuclear Medicine Radiation Dose History. Medscape Radiology. Accessed 06-05-2014:

4. Lin, Eugene C. Radiation Risk From Medical Imaging. Mayo Clinic Proceedings , Volume 85 , Issue 12 , 1142 - 1146

5. Pierce, DA and Preston, DL. Radiation-induced cancer risks at low doses among atomic bomb survivors. Radiat Res. 2000; 154: 178–186

6. Smith-Bindman, R, Lipson, J, Marcus, R et al. Radiation dose associated with common computed tomography exams and the associated lifetime attributed risk of cancer. Arch Intern Med. 2009; 169: 2078–2086

7. Silva, AC, Lawder, HJ, Hara, A et al. Innovations in CT dose reduction strategy: application of the adaptive statistical iterative reconstruction algorithm. AJR Am J Roentgenol. 2010; 194: 191–199

8. Tubiana, M, Feinendegen, LE, Yang, C, and Kaminski, JM. The linear no-threshold relationship is inconsistent with radiation biologic and experimental data. Radiology. 2009; 251: 13–22

9. Verdun, FR, Bochud, F, Gudinchet, F et al. Radiation risk: what you should know to tell your patient. Radiographics. 2008; 28: 1807–1816

Submitted by Salibian, Raffi