Sample
The analysis of the routine head CT studies was supported on a sample composed by 115 patients (59 female,
59 male) with ages between 18 and 94 years old mostly coming from the urgency room (109 patients).
As presented in Table 2 ,
in this sample four different scan protocols were used,
but more frequently Protocol I and Protocol II.
Descriptive statistical analysis
Principal descriptive results obtained were resumed in the Table 3 .
Concerning total DLP variable an average of 1109,82 ± 9,20 mGy.cm was assessed and a significant statistical difference was observed for groups gender and protocol.
Concerning CTDI values,
in the posterior fossa an average of 96,70 ± 0,52 mGy was assessed with significant statistical difference between Protocol I and II.
At the Supratentorial Fossa range CTDI average was 59,54 ± 0,28 mGy.
Gantry tilt average obtained was 13,14 ± 0,89° and showed a positive correlation with age.
Correlations and Linear regressions
In Table 4 correlations were briefly presented.
As expected,
total DLP presented a stronger correlation with posterior fossa DLP than with supratentorial DLP.
No correlation was observed between gantry tilt and DLP variables for the significant level used in this study.
A weak correlation between gantry tilt and scan angle was observed,
but as shown in Fig. 3,
even there is no statistical linear regression and a diffuse presentation in the relation between gantry tilt and scan angle,
there is a growing trend to gantry tilt values when scan angle increases to zero.
With respect to scan angle,
it was correlated with the dimension of the study ranges and consequently with the DLP variables.
It was observed a strong correlation between scan angle and DLP variables,
stronger for the posterior fossa than for the supratentorial range.
Linear Regressions were drawn and statistical linear regressions were observed between scan angle and Total DLP and posterior fossa DLP (Fig. 4).
Same results were observed when the sample was divided into groups (sex and protocol),
but with stronger R-squares (see Table 5).
Phantom Assay
As expected,
the study phantom showed a perfect linear regression between gantry tilt and scan angle in relation with the standard positioning of the phantom to all the exposures.
Also the linear regressions drawn between scan angle and DLP variables shown very strong linear regressions with very high R-squares (Fig. 5).
Finally,
5-degrees scan angles intervals were analysed and results were presented in Fig. 6 As observed,
-5° to 0° was the interval where lower dose (lower DLP total) was used.