Evaluation of an a priori scatter correction algorithm for cone-beam computed tomography based range and dose calculations in proton therapy

Abstract

Background and purpose: Scatter correction of cone-beam computed tomography (CBCT) projections may enable accurate online dose-delivery estimations in photon and proton-based radiotherapy. This study aimed to evaluate the impact of scatter correction in CBCT-based proton range/dose calculations, in scans acquired in both proton and photon gantries. Material and methods: CBCT projections of a Catphan and an Alderson phantom were acquired on both a proton and a photon gantry. The scatter corrected CBCTs (corrCBCTs) and the clinical reconstructions (stdCBCTs) were compared against CTs rigidly registered to the CBCTs (rigidCTs). The CBCTs of the Catphan phantom were segmented by materials for CT number analysis. Water equivalent path length (WEPL) maps were calculated through the Alderson phantom while proton plans optimized on the rigidCT and recalculated on all CBCTs were compared in a gamma analysis. Results: In medium and high-density materials, the corrCBCT CT numbers were much closer to those of the rigidCT than the stdCBCTs. E.g. in the 50% bone segmentations the differences were reduced from above 300 HU (with stdCBCT) to around 60-70 HU (with corrCBCT). Differences in WEPL from the rigidCT were typically well below 5 mm for the corrCBCTs, compared to well above 10 mm for the stdCBCTs with the largest deviations in the head and thorax regions. Gamma pass rates (2%/2mm) when comparing CBCT-based dose re-calculations to rigidCT calculations were improved from around 80% (with stdCBCT) to mostly above 90% (with corrCBCT). Conclusion: Scatter correction leads to substantial artefact reductions, improving accuracy of CBCT-based proton range/dose calculations.

Impact of interfractional target motion in locally advanced cervical cancer patients treated with spot scanning proton therapy using an internal target volume strategy

“…A plan of the day concept as proposed by van de Schoot et al [2] or a real-time re-planning strategy as reported by Jagt et al [11] for cervix cancer with a prior plan-library may further improve target coverage and OARs sparing for patients subject to large interfractional anatomical variations. Nevertheless, the workflow of the latter strategy is still not clinically available and some challenges remain to be addressed before implementation of on-line adaptive RT/PT in clinical routine: improved image quality and automatic contouring would be required for this approach to become feasible and cost-effective [23] , [24] .…”

Planning CT‐guided robust and fast cone‐beam CT scatter correction using a local filtration technique

“…Using pCT as prior information for CBCT scatter correction has demonstrated effectiveness. The improved CBCT may be directly used for adaptive treatment planning 31,45–49 . However, pCT and CBCT are acquired at different time with different anatomy.…”

“…The improved CBCT may be directly used for adaptive treatment planning. 31,[45][46][47][48][49] However, pCT and CBCT are acquired at different time with different anatomy.The performance of the existing methods heavily relied on the registration accuracy. The proposed LF-RR method does not rely on the registration accuracy, hence provides an alternative and efficient scatter correction pathway.…”

“…The pCT‐based scatter correction methods 29–33 combine the CT number information in pCT with anatomy information in CBCT, enabling accurate scatter correction without extra hardware modification and sophisticated physical modeling. However, the performance of the pCT‐based method heavily relies on the consistency of the pCT acquired previously and the CBCT under reconstruction 31 . It is difficult, if not impossible, to achieve a decent registration because of the day‐to‐day deformation of the patient's anatomy.…”

Deep learning methods for enhancing cone‐beam CT image quality toward adaptive radiation therapy: A systematic review

“…Alternatively, the use of unpaired data with cycle-GAN mitigates the need to perform any smoothing operations during a priori correction, allowing the network to learn both low and high frequency sources of intensity errors-a characteristic not possible under the original a priori implementation. [28][29][30]83 The use of DL techniques for CT reconstruction has predominantly involved sparse-view and low-dose acquisitions as there is no ideal tomographic ground truth for learning strategies. [105][106][107][108][109][110] Iterative reconstruction (IR) applications utilizing DL can be categorized into "plug-and-play" and "unrolling" methods.…”