In this preliminary study,
we investigated the reliability of trained non-radiologist raters for the evaluation of acute non-contrast CT-scans in patients with TBI,
using a digital structured reporting tool based on the most recent CDE templates.
Our findings indicate that non-radiologist raters,
with a background in neuroscience,
and given adequate training,
can reliably and consistently generate structured reports in patients with TBI.
Caution is,
however,
advised,
and we recommend that a trained (neuro)radiologist,
with many years of experience,
should check the reports.
This result is not unexpected.
Similar results have been found for trained neuroradiographers [15].
Some centers are even advocating to allow these trained non-radiologists to provisionally report head CT’s on-call.
Of course,
then checked by a radiologist,
similar to the supervision process used for radiology trainees [15].
Our study shows that structured reporting can succesfully be implemented for patients with TBI,
certainly in a research setting.
There is a lot of interest throughout the radiological community to apply uniform terminology for reporting on patients with TBI.
As an example,
the American College of Radiology (ACR) is currently developing the Head Injury Imaging Reporting and Data System (HI-RADS),
in order to standardize the reporting and data collection of imaging in patients with TBI,
building on the CDEs [16].
The purpose of this initiative is to reduce variability and errors in imaging interpretation,
to improve communication with referring physicians,
to facilitate quality assurance and research and,
in the long run,
to improve patient outcome [16].
In conclusion,
trained non-radiologist raters,
with a background in neuroscience/ neuroanatomy,
can reliably and accurately identify basic CT-scan features of TBI,
and enter structural abnormalities into research databases via structured reporting.
Our data indicate that they can achieve substantial to almost perfect agreement with expert neuroradiologists.