dc.description.abstract |
Introduction
Computed Radiography (CR) is currently the main leading digital radiography system
that was introduced to interchange from conventional film-screen radiography systems.
The key advantage of CR over film-screen radiography is the ability to perform postprocessing,
which allows image recording faults to be rectified. This, inevitably leads
one to inquire whether or not the required radiographic techniques are still being
employed prior to post-processing when CR systems are used.
Aim of the study
The aim of this study was to assess and possibly enhance image recording techniques
employed when using computed radiography imaging systems in private and
government hospitals in the Eastern Cape province, South Africa.
Methodology
A retrospective study design, using a self-designed checklist, was utilised to assess
the image recording techniques used by diagnostic radiographers when producing CR
images. The checklist consists of quantitative data with qualitative elements. Images
of the chest and abdomen were evaluated by the researcher and two other assessors.
The team assessed a total of 720 (PA/AP and LAT) chest and (erect and supine)
abdominal images individually, which were copied from the CR workstations for each
assessor.
The data were categorically captured by the researcher and analysed by the quality
assurance (QA) radiographer to ensure accuracy before sending the hard and soft
copies of the sample to the statistician. Thus, the data were also provided to the
statistician to verify the accuracy of the checklist results copied into an Excel
spreadsheet, through the use of a data theme analyses technique. The technique that
was used in the analysis was that of key-words-in-context. Descriptive data, namely
frequencies and percentages, were calculated for categorical data. Means and
standard deviations or medians and percentiles were calculated from the numerical
data obtained.
Results
Image recording techniques assessed with a consistently high level of accuracy were
‘part selection on CR workstations’, ‘gridline artefact’ exclusion and ‘CR scanner
malfunction’, which resulted in averages exceeding 95%. The results also indicated
that an unacceptably high number of chest and abdominal examinations that were
assessed had averages of non-optimal positioning: [chest (41%) and (abdomen 41%)],
non-anatomical markers: [chest (73%) and abdomen (59%)], and no collimation
applied: [chest (64%) and abdomen (72%)]. The most noticeable assessments relating
to artefacts were foreign objects on the patients manifesting in PA/AP chest (14%),
LAT chest (20%), erect abdominal (23%) and supine abdomen (13%) images. Radiographers measure their processed images through EI values, which were
assessed as either ‘over exposed’, ‘under exposed’ or ‘in range’. The EI results
indicated that the majority of PA/AP chest was ‘in range’ (43%), whereas LAT chest
was ‘under exposed’ (45%). Conversely, the abdominal images only showed
underexposure as an average of 15% (18% for erect and 11% for supine abdominal
images). It is striking to note that overexposure occurred in 52% of the abdominal
images compared to 15% overexposure in chest images. Observed with assessment,
histogram errors occurred in 7% LAT chest to 5% PA/AP chest, whereas erect and
supine abdominal images had equivalent average histogram errors of 3% (n=16) each.
Looking at the image quality assessment of all chest and abdominal images,
satisfactory results relating to distortion, noise level and the degree of sharpness
occurred. However, the study did identify that contrast and density technique image
quality falling below an acceptable ‘3’ qualifier value could be improved in both chest
and abdominal images assessed.
Conclusion and recommendations
All research questions and objectives of the study were addressed. This enabled the
researcher to conclude that the three key areas requiring attention were: (i)
radiographic practice, (ii) setting the exposure and (iii) avoiding artefacts through
practical techniques. Recommendations are made to address these findings. Four
sections, namely functional, technical, practical, and quality assurance
recommendations are proposed. |
en_US |