Streamlining custom bolus fabrication via 3D-to-2D unfolding using spectral mesh flattening

Accurate fabrication of custom boluses is essential in radiation therapy to ensure optimal dose delivery to superficial tumors, particularly in anatomically complex regions. This study presents a novel method that utilizes spectral mesh flattening (SMF) to unfold three-dimensional (3D) virtual bolus designs into two-dimensional (2D) contours, aiming to enhance the fabrication workflow and improve reproducibility in bolus shape and placement. Using computed tomography (CT) scans of a phantom with radiopaque wires delineating target regions such as the nose, chin, and ear, virtual boluses of 0.5 cm thickness were designed within a treatment planning system. The 3D mesh geometries of these boluses were then exported and processed using a custom-developed software tool, ONCOFLAT, which implements the SMF algorithm to generate 2D representations while minimizing geometric distortion. These 2D contours were printed and used as cutting guides for the fabrication of flat bolus materials. After fabrication, the boluses were applied to the phantom and rescanned, and their accuracy was assessed by comparing the physical boluses to the original virtual designs using the Dice Similarity Coefficient (DSC) and the Hausdorff distance. The SMF algorithm successfully unfolded complex 3D geometries into 2D contours, and the ONCOFLAT software enabled a streamlined process that reduced the total design-to-fabrication time to under five minutes. The fabricated boluses closely conformed to the intended anatomical surfaces, with DSC values ranging from 0.59 to 0.62 and average Hausdorff distances below 1.3 mm. The 95% Hausdorff distances ranged from 3.50 mm to 4.22 mm. These results demonstrate that the integration of SMF within ONCOFLAT offers a fast, reproducible method for fabricating patient-specific boluses for complex anatomy. The approach shows promise in improving the consistency and effectiveness of dose delivery in radiation therapy, representing a meaningful advancement in personalized treatment planning.