Not checkedNot checked for notices yet.
Error analysis of 4-dimensional computed tomography image reconstruction in uniform motion state
DAI Hongya, ZHOU Yibing, QIAN Jindong and 4 more
Di-san junyi daxue xuebao · 2021
doi:10.16016/j.1000-5404.202104025 (opens in a new tab)
We email you when a new notice or a question about this paper appears.
Abstract
Objective To investigate the accuracy of 4-dimensional computed tomography (4DCT) image reconstruction in uniform motion state using respiratory motion phantom. Methods The cosine wave movements of lung tumor in the craniocaudal direction were simulated on a respiratory motion phantom, and the simulations were divided into 6 groups according to the respiratory motion amplitude (A) and the breathing period (T), that is, A=0.5 cm, T=6 s; A=0.5 cm, T=3 s; A=1.0 cm, T=6 s; A=1.0 cm, T=3 s; A=1.5 cm, T=6 s; A=1.5 cm T=3 s. 4DCT scans were acquired using Philips big bore CT simulator, and the movements were scanned twice for each group, with same scanning parameters. Finally, the internal target volume (ITV) was delineated on the reconstructed 4DCT images, and the volume and length of ITVs were measured and compared with their theoretical values to analyze reconstruction errors of 4DCT. Results There were errors in all the 6 groups of ITV delineated on the reconstructed 4DCT images. The relative error range of ITV volume was from 1.13% to 27.64%, and that of ITV length was from -0.95~0.03 cm. The average ITV length error was 0.25 cm. When A=1.5 cm and T=6 s, the reconstruction error showed the maximum, and the volume errors presented 27.64% and 19.04%, while the length errors were 0.11 and 0.84 cm between the 2 scans. There were also differences between the results of 2 scans under the same respiratory motion parameters and scanning parameters. When A=0.5 cm and T=6 s, the volume errors of ITV were 14.47% and 4.43% in the 2 scans, with a difference of 10.00%. Conclusion There remain errors in 4DCT image reconstruction under uniform motion, and the size of the error is randomized and unpredictable.
Abstract licence: CC BY
Reads
Read by people on this page
None yet.
Checked or read by AI through naulon
None yet.
A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.
Are you an author?
Is this your paper?
Claim it to add your note and be paid when agents cite it. Free.