Software tracks CT scan radiation exposure

It is hoped that the software will arm radiologists, medical physicists and patients with more accurate data for making informed decisions about the potential risks and benefits of CT scan procedures.

Project leader Prof Xie George Xu of the Rensselaer Polytechnic Institute said: ‘The radiation exposure from a single CT scan is still relatively small when compared with the clinical benefit of the procedure, but patients often receive multiple scans during the course of their diagnostic or therapeutic procedure. Our new software should help to record the exposures more accurately and more consistently.’

A recent report by the US National Council on Radiation Protection and Measurements (NCRP) detailed how the US population is now exposed to seven times more radiation every year from medical imaging exams than it was in 1980.

While CT scans only account for 10 per cent of diagnostic radiological exams, the procedure contributes disproportionately — about 67 per cent — to the collective medical radiation exposure.

To help mitigate this risk, several bodies have called for the establishment of a centralised, patient-specific ‘dose registry’ system. Such a system would track over time the amount of CT scans a patient undergoes, and the radiation exposure resulting from those procedures. However, current software packages for tracking CT scan radiation exposure have fundamental limitations and are insufficient for such a critical task, according to Xu.

The new software Xu and his team are developing — VirtualDose — takes into consideration a patient’s individual characteristics, including age, sex, pregnancy, height and weight. By entering these data into the software, the program creates a virtual 3D phantom.

These anatomically realistic phantoms accurately model the patient’s internal organs, and detail how radiation interacts with each organ. The phantom, in turn, allows physicians and researchers to compare the levels of radiation exposure a patient gets from different CT scanning protocols or different scanner designs.

Current software for CT radiation dose reporting uses outdated models of patients, and often lacks necessary software features, Xu said. This makes it nearly impossible to accurately track and record radiation exposure to organs from X-rays.

Xu added that personalised virtual phantoms are particularly important for predicting radiation exposure from CT scans for the groups most sensitive to radiation: children and pregnant women. These groups are ignored by nearly all dose-measurement software.

Clinical testing of the software will take place at several hospitals, including Massachusetts General Hospital and Shands Hospital at the University of Florida.