Methods and materials:
CT exposure data of ~24250 patients,
collected between 2013 and 2018 by 9 medical physicists in more than 150 public and private French hospitals,
were retrospectively analysed.
Over-ranging was calculated as the difference between total exposed scan length and total imaged scan length ( Fig. 1 ).
Per each scanner,
mean over-ranging ± standard deviations were calculated.
Inclusion criteria were the availability of slice thickness,
collimation and pitch.
Over-ranging in fourteen different CT models from GE,
Philips and Siemens were compared.
The impact of collimation and pitch on over-ranging was also studied.
Over-ranging values inherent to CT scanners with ~2 cm and ~4 cm total collimation were compared.
Additionally,
for each manufacturer and scanner model,
the efficiency over time of solutions introduced for the management and limitation of over-ranging on latest CT models was investigated.
For each manufacturer and a given total collimation (~2 or ~4 cm),
data from different scanner models where pooled together when identical over-ranging values were obtained for a given pitch-collimation combination.
This leads to the following merging:
- GE Optima CT 520 / 540 / 580
- GE Optima CT 660 (before 2016)
- GE Optima CT 660 from 2016 and Revolution Evo
- Philips Ingenuity Flex 32
- Philips Ingenuity Core 64 / Core 128 / Elite / CT
- Siemens Somatom Definition AS64
- Siemens Somatom Definition AS+ / Edge / Flash
Results:
In this study,
74% of analysed data were collected from private hospitals and 26% from public institutions.
A total of 155 CT scanners were included,
their distribution per manufacturer and per model type is given in Fig. 2 .
Over-ranging (Lov) for each CT protocol collimation and pitch were plotted for ~2 and ~4 cm total collimation CT scanners (respectively Fig. 3 and Fig. 4 ).
When individually studying the impact of collimation,
it was found that increasing collimation from 2 to 4 cm would yield a two-folds increase in over-ranging for an identical manufacturer and equipment model.
Meanwhile,
for two ~4 cm maximum collimation CT scanners of Siemens and Philips,
an over-ranging value up to 5.9 times higher was found for Philips although identical collimation (~4 cm) and pitch (0.45) were considered.
Additionally,
for GE Optima CT 660 scanners,
a 20-25% reduction of over-ranging was observed before and from the year 2016 (Dynamic z-axis tracking technology).
Since each CT scanner manufacturer use collimation and pitch in a different way to achieve high quality images (Siemens and Philips consider a similar approach),
typical CT protocols for head,
chest,
abdomen-pelvis and lumbar acquisitions were collected to compute over-ranging.
This comparison was done for the three most widely used CT scanners in our sample: GE Optima CT 660,
Philips Ingenuity Core 64 and Siemens Somatom Definition AS+ scanners.
Fig. 5 shows that,
for all clinical applications,
lowest over-ranging values can be obtained on Siemens Somatom Definition AS+ although this model involves a 3.84 cm total collimation.
In opposition,
Philips Ingenuity Core 64 was the model to induce largest over-ranging on all anatomical sites.
In the specific case of head CT,
the computed over-ranging was similar for Optima CT 660 and Siemens Definition AS+ although the former involved a 2 cm total collimation vs.
4 cm on the latter scanner.
The largest registered difference in over-ranging was of 5.3 cm for abdomen-pelvis CT acquisition (Philips Ingenuity Core 64: collimation 4 cm,
pitch 0.984,
Lov 7.4 cm vs.
Siemens Somatom Definition AS+: collimation 3.84 cm,
pitch 0.75,
Lov 2.1 cm).