ESTRO 2024 - Abstract Book

S3836

Physics - Image acquisition and processing

ESTRO 2024

Purpose/Objective:

In radiotherapy, simulation CT scans are usually performed at a standard tube voltage of 120 kV. This choice simplifies the planning workflow since only one calibration curve is needed. The Balearic Islands hospital has recently acquired a Siemens SOMATOM CT scanner with the DirectDensity® algorithm [1], which opens up the possibility of using a single calibration curve for various voltages. This algorithm creates scope for tube voltage adaptation (e.g., CARE kV), which is routinely used in diagnostic radiology to improve image quality and reduce radiation exposure. After 6 months of clinical use, we study the automatic choice of tube voltage and compare the dose levels with regards to fixed-voltage protocols.

Material/Methods:

A total of 327 simulation CT scans have been included in this survey, performed between November 2022 and May 2023, of various locations: breast (163), pelvis (96), lung (44), and head/neck (24). For each protocol, median values of CTDI vol and DLP were calculated and compared to published reference levels in radiotherapy, which use fixed-voltage protocols [2]. We have also analyzed the automatic choice of tube voltage per site.

Results:

Median dose levels for breast, lung, and pelvis were well below the UK national reference levels for simulation CT (see table below, reference values from [2]). The most significant reduction is observed in lung scans, with a reduction of 36% in CTDI vol . Dose levels were very similar for head/neck (CTDI vol for HN is based on the 16 cm phantom, while the others on the 32 cm).

CTDI vol (mGy)

DLP (mGy·cm)

Ref. CTDI vol

Ref. DLP

Breast

8

310

10

390

Pelvis

11

444

16

610

Lung

9

310

14

550

Head/Neck

45

2160

49

2150

With regards to tube voltage, the highest variability results from breast scans: 50% of them are performed at 80–90 kV, while only 5% at 120 kV. A large number of scans (25%) were performed at 140 kV. Pelvic scans are preferentially performed at 120 kV (75%), while head/neck are all completed at 120 kV. We observed that many of the breast scans were performed at 140 kV due to the presence of high-density elements. In breast simulation scans, the arms are bent, which increases the attenuation in the head/neck area. The situation gets worse for patients who cannot stretch out their arms, or who have dental implants. In those cases, the scan is performed at an unnecessarily high tube voltage, which can reduce image quality in the area of interest (i.e., chest).