Influence of CT automatic tube current modulation on uncertainty in effective dose.

Computed tomography (CT) scanners are equipped with automatic tube current modulation (ATCM) systems that adjust the current to compensate for variations in patient attenuation. CT dosimetry variables are not defined for ATCM situations and, thus, only the averaged values are displayed and analysed. The patient effective dose (E), which is derived from a weighted sum of organ equivalent doses, will be modified by the ATCM. Values for E for chest-abdomen-pelvis CT scans have been calculated using the ImPACT spreadsheet for patients on five CT scanners. Values for E resulting from the z-axis modulation under ATCM have been compared with results assessed using the same effective mAs values with constant tube currents. Mean values for E under ATCM were within ±10 % of those for fixed tube currents for all scanners. Cumulative dose distributions under ATCM have been simulated for two patient scans using single-slice dose profiles measured in elliptical and cylindrical phantoms on one scanner. Contributions to the effective dose from organs in the upper thorax under ATCM are 30-35 % lower for superficial tissues (e.g. breast) and 15-20 % lower for deeper organs (e.g. lungs). The effect on doses to organs in the abdomen depends on body shape, and they can be 10-22 % higher for larger patients. Results indicate that scan dosimetry parameters, dose-length product and effective mAs averaged over the whole scan can provide an assessment in terms of E that is sufficiently accurate to quantify relative risk for routine patient exposures under ATCM.

Relationships between patient size, dose and image noise under automatic tube current modulation systems.

Automatic tube current modulation (ATCM) systems are now used for the majority of CT scans. The principles of ATCM operation are different in CT scanners from different manufacturers. Toshiba and GE scanners base the current modulation on a target noise setting, while Philips and Siemens scanners use reference image and reference mAs concepts respectively. Knowledge of the relationships between patient size, dose and image noise are important for CT patient dose optimisation. In this study, the CT patient doses were surveyed for 14 CT scanners from four different CT scanner manufacturers. The patient cross sectional area, the tube current modulation and the image noise from the CT images were analysed using in-house software. The Toshiba and GE scanner results showed that noise levels are relatively constant but tube currents are dependent on patient size. As a result of this there is a wide range in tube current values across different patient sizes, and doses for large patients are significantly higher in these scanners. In contrast, in the Philips and Siemens scanners, tube currents are less dependent on patient size, the range in tube current is narrower, and the doses for larger patients are not as high. Image noise is more dependent on the patient size.

Comparison of different phantom designs for CT scanner automatic tube current modulation system tests.

Modern CT scanners modulate tube current during scans according to patient size, shape and attenuation. However, the ATCM (automatic tube current modulation) systems for different CT manufacturers work on different principles. Although the systems are used for the majority of patients and examinations, there is no standard phantom for routine quality control of CT scanner ATCM operation. The ideal phantom for testing these systems should be capable of evaluating how tube current and image quality as well as dose vary according to changes in patient size and shape. For this study, a conical phantom designed by ImPACT has been compared with two phantoms made from elliptical sections with varying dimensions. The concept of the designs is to reflect the ATCM performance for the varying shapes and dimensions along the length of the human body. The first phantom comprises five elliptical sections with a wide range of different dimensions and the second has three sections that are more similar in size. The phantoms have been used to test ATCM systems for Philips, Siemens, GE and Toshiba scanners. Although the results of the tube current modulation patterns were similar for all CT scanners, the abrupt changes in attenuation for the first sectional phantom provoked an abnormal ATCM response for the GE and Toshiba scanners. The second sectional phantom was developed from the results of the first, and was more effective for ATCM system testing and could be used for dose and image quality assessment in standard positions. However, the ImPACT conical phantom provided the best overall assessment of performance in terms of tube current modulations and noise pattern.

A study of CT dose distribution in an elliptical phantom and the influence of automatic tube current modulation in the x-y plane.

Computed tomography (CT) performance assessments relating to patient dose to the body are made conventionally in 320 mm diameter cylindrical acrylic phantoms. The cross section of the human trunk is closer to an ellipse and automatic tube current modulation (ATCM) systems adjust the exposure level with orientation in the x-y plane, changing the dose distribution within the body. This study has investigated differences in the distributions of dose within a standard cylindrical body phantom and an elliptical dosimetry phantom for Toshiba, General Electric and Philips CT scanners, and recorded changes with the application of the ATCM. Single slice dose profiles have been recorded within the phantoms using Gafchromic film. CT dose indices along 100 mm lengths have been calculated and data sets combined to simulate helical scans, from which values for cumulative doses have been derived. The doses in the centre of the elliptical phantom are 70-100% larger than for the cylindrical one and in the anterior are around 20-40% larger, while the doses in the lateral positions are similar for the two phantom shapes. The differences between the anterior and lateral doses were larger for the Toshiba scanner and this is thought to be linked to the narrower profile of the beam produced by the bow-tie filter. When the ATCM mode for the Toshiba scanner is implemented, the doses in the anterior and posterior positions are reduced preferentially, bringing them closer to the doses in the lateral positions.

Application of Gafchromic film in the study of dosimetry methods in CT phantoms.

Gafchromic film has been used for measurement of computed tomography (CT) dose distributions within phantoms. The film was calibrated in the beam from a superficial therapy unit and the accuracy confirmed by comparison with measurements with a 20 mm long ionisation chamber. The results have been used to investigate approaches to CT dosimetry. Dose profiles were recorded within standard CT head and body phantoms and scatter tail data fitted to exponential functions and extrapolated to predict dose levels in longer phantoms. The data have been used to simulate both CT dose index (CTDI) measurements with ionisation chambers of differing length and measurements of cumulative doses with a 20 mm chamber for scans of varying length. The results show that the length of a pencil ionisation chamber is the most significant factor affecting measurements of weighted CTDI (CTDI(w)) and a 100 mm chamber would record 50-61% of the dose measured with a 450 mm one. The cumulative dose measured at the centre of a 150 mm long body phantom records over 70% of the equilibrium dose from a helical scan of a longer phantom. For routine CT dosimetry tests, the determination of correction factors could allow measurements with a 100 mm chamber to be used to derive the CTDI that would be recorded with a longer chamber, and cumulative doses measured with a 20 mm chamber in shorter phantoms to be used to calculate equilibrium doses for helical scans.

Scattering and attenuation correction in emission tomography in the nuclear industry.

A multi-energetic gamma-ray source has been used in conjunction with titanium and water-bath phantoms to simulate the scattering and attenuation effects encountered in nuclear industrial applications. The effect of scattered photons on the image reconstructions has been investigated by the subtraction of varying fractions of the scattered counts from the photopeak counts. A dual energy attenuation correction has also been applied to the data. In order to evaluate the effects of these corrections a "fidelity" factor has been calculated.