The aim of this study is to compare patient geometrical uncertainties in the treatment of breast boost three-dimensional conformal radiation therapy (3D-CRT) considering both manual alignment and automatic different registration methods in cone-beam computed tomography (CBCT).

A total of 85 patients were chosen for this study. A total of 254 registrations of CBCT vs planning computed tomography (CT) were retrospectively performed using automatic registration algorithms from Elekta XVI system (Clipbox and Mask) to detect patient setup uncertainties. All CBCTs were also matched manually by three health professionals. Mean shift values obtained with manual registration performed by health professionals were used as reference. Absolute value of difference between automatic algorithm shifts and reference values shifts was collected for each enrolled patient considering the three different spatial directions (x,y,z), and the magnitude was calculated (δm for Mask and δc for Clipbox).

Data analysis showed a significant difference in δm and δc. t-Test statistics showed a high difference between Mask and Clipbox, in particular mean δm = (1.3 ± 0.1) mm and δc = (3.3 ± 1.2) mm (p-value <0.0001). Mask algorithm was performed in a very similar way with respect to the reference alignment, and the differences between these two procedures were of the order of 1 mm. Clipbox algorithm showed larger differences with manual registration.

These results suggest that the Mask algorithm may be the optimal choice for patient setup verification in clinical practice for breast boost treatment in 3D-CRT.