Abstract

Impact of CT Window Contouring on Breast Plans

Objective: To assess the impact of CT window for postmastectomy breast irradiation on Gross tumor volume (GTV) and Internal target volume (ITV) definitions for a target with large movements. Materials and Methods: Tumor movement was simulated with a phantom composed of a water filled table tennis ball. The phantom was elliptically moved in the ap-pa and cranio-caudal or in ap-pa and lateral direction, respectively. CT scans were acquired for the ball with and without movement, both in a “slow” 3D-CT scan and in a 4D-CT scan. The treatment protocol used involved a boost block with ipsilateral breast exclusive of 4 cm around the breast plug. The ball was contoured as GTV or ITV using different window settings (for e.g. “breast window”, “soft tissue” window) and the volumes were compared to mathematically calculated volumes (“gold standard”). Results: In the 3D-CT of the ghost without development window settings had just a minor effect on GTV. The maximal deviation to the figured size was 12%. Amid moderate 3D-CT sweeps of the moving ball extensive contrasts between the window, settings were found. Best concordance to the ascertained qualities demonstrated the breast window setting (-16.8% and 19.1%). Unmistakable thought little of ITVs were formed utilizing the 500 and delicate tissue window (-84% to -99%). A significant littler impact was found for the 4D-CT. Movement heading had a vital effect on the objective outline, with an underestimation of volumes in the cranio-caudal bearing and an overestimation of volume in the horizontal course for the moderate 3D-CT shaped in breast window. Conclusions: Window settings are vital in the shaping of moving targets, while they are less pertinent for static targets. In moving targets cautious thought of window settings for shaping ought to be finished. Regardless of whether breast windowing is utilized, we emphatically prescribe reconsideration of the formed focus in various planes for the moderate 3D-CT. 4D-CT limits vulnerabilities.

• PDF