Simulation of the discrepancy in 4DCT-based cumulative dose using digital phantoms / 中华放射肿瘤学杂志

ABSTRACT

Objective To conduct a computer simulation to evaluate the discrepancy between the cumulative doses calculated by four-dimensional computed tomography (4DCT) images and 4DCT scans (for real-time respiratory motions) due to the patient's irregular breathing.Methods A series of digital phantoms were generated from a patient's 4DCT images to simulate 4DCT images and 4DCT scans (for real-time respiratory motions) resulting from various irregular breathing curves.A six-field intensity-modulated radiotherapy plan was generated.Two cumulative doses in the target were calculated.The first one, named Dall, was calculated by tracking the point displacements in the target manifested on the 4DCT images;the second one, named D4D, was calculated based on the point displacements along the whole breathing motion during 4DCT scanning.Dose discrepancy between D4D and Dall was calculated to evaluate the correlation between breathing pattern and dose discrepancy in the target.Results The dose discrepancy in the target was correlated with mean motion excursion and the standard deviation of motion excursion.ΔDmin(ΔD99) in the target increased from 2.39%(2.04%) to 11.91%(5.24%) as the mean motion excursion increased from 5 mm to 15 mm, and increased from 5.93%(2.15%) to 14.65%(5.01%) as the standard deviation of motion excursion increased from 15% to 45% of the mean motion excursion.When the mean period increased from 3 s to 5 s or the standard deviation of period increased from 10% to 40% of the mean period,ΔDmin(ΔD99) in the target was greater than 6.0%(2.0%), but less than 9.0%(3.0%).When the target diameter was 2 cm, 3 cm, and 4 cm,ΔDminΔD99) in the target was 11.88%(5.50%), 6.91%(2.42%), and 7.53%(3.62%), respectively.Conclusions There is a large discrepancy between the cumulative doses calculated using 4DCT images and 4DCT scans (for real-time respiratory motions) when the patient has irregular breathing.This dose discrepancy depends on mean motion excursion and the standard deviation of motion excursion, but has little relationship with mean period, the standard deviation of period, and tumor volume.