Comparative evaluation of image registration methods with different interest regions in lung cancer radiotherapy

Abstract

Background: Lung cancer is a leading cause of morbidity and mortality worldwide. Radiotherapy for lung cancer is beneficial in both the radical and palliative settings, and technologic advances in recent years now afford an opportunity for this treatment to be more targeted than ever before. Although the delivery of more accurate forms of radiotherapy has minimized the risks of side-effects, how to utilize this treatment to optimize outcomes remains questionable. This study aimed to evaluate the accuracy of cone beam computed tomography (CBCT) image registration used in image-guided radiotherapy, providing reasonable guidance for clinic application of CBCT in lung cancer. Methods: A total of 53 patients with lung carcinoma including 34 central and 19 peripheral lesions were collected in this study. Varian-IX linear accelerator on-board imaging (OBI) system was used to acquire CBCT scans in threedimensional (3D) conformal radiotherapy before delivery. Different regions (whole lung/target/vertebrae/ipsilateral structure) were manually registered, and the position deviation and the registration time were analyzed. Results: It was suggested that 34 cases belonged to central type and 19 cases belonged to peripheral type. The volume of left lung and right lung was 1242.98 ± 452.46 cc, 1689.69 ± 574.31 cc, respectively. Tumor size was 6.65 ± 3.87 cm in diameter, and 129.67 ± 136.48 cc in volume. The percentage of left lung and right lung was 6.17 ± 1.24%, 4.74 ± 0.38%, respectively. The position deviation value and absolute value of image registration methods of X, Y and Z axis were not significant (P > 0.05). However, registration time (s) between whole lung registration group, tumor registration group, vertebral body registration group, affected lung registration group, and artificial registration group, was 3.651 ± 0.867 s, 1.144 ± 0.129 s, 1.226 ± 0.126 s, 2.081 ± 0.427 s, 179.491 ± 71.975 s, respectively. The differences were significant (P < 0.05). The registration differences between small tumor group and large tumor group were not statistically significant (P > 0.05). Conclusion: The automatic image matching of OBI is accuracy and high reliability in recognition of offset error. Registering body or ipsilateral structure is recommended to be used in CBCT for lung cancer.

Value of virtual bronchoscopic navigation and transbronchial ultrasound-guided sheath-guided exploration in diagnosis of peripheral lung cancer

“…Early peripheral lung cancer has no significant symptoms, and when the symptoms are obvious, the cancer is in a late stage. Globally, lung cancer has a high incidence and mortality rate[ 5 , 6 ]. If patients with peripheral lung cancer have advanced metastasis, it is necessary to have a clear molecular diagnosis based on different targets to initiate individualized treatment and promote a therapeutic effect.…”

Comparative analysis of setup margin calculation in cone beam CT, by van Herk formula, using two different image registration methods

“…The van Herk formula used CBCT image shift measurements to calculate the PTV margin. 6 In head-neck VMAT treatment, there were very few published literatures comparing BL and ST matching with respect to possible impact in PTV margin. The current study analysed translational errors in setup adjustments utilising CBCT images in head-neck cancer treatment using VMAT and two registration approaches.…”

“…With a high gradient in the distribution, VMAT demands a proper setup through image guidance to avoid missing any clinical or subclinical disease, leading to locoregional failure. 4 Although there were limited solutions for image guidance, like two-dimensional portal imaging, stereoscopic imaging and surface guidance, the most accurate and the standard was the linear accelerator embedded in cone beam CT. 5,6 Quite a few articles showed the high accuracy of cone beam computed tomography (CBCT) and compatibility with other imaging systems.…”

“…Cao et al compared image registration methodology with different ROIs in lung cancer radiotherapy. 6 It was found that the automatic image matching of onboard imaging (OBI, Varian, Palo, Alto, CA) was accurate and highly reliable for detecting offset errors.…”

Evaluation of an artificial intelligence U-net algorithm for pulmonary nodule tracking on chest computed tomography images

“…In recent years, various algorithms that use deep learning have been studied. 3 While lung image registration studies have included several that were targeted at inhalation and exhalation 4 and radiation therapy, 5 as far as we are aware, none have applied image registration to the follow-up of lung nodules. Therefore, the purpose of this present study was to apply image registration for the follow-up of lung nodules and to verify the feasibility of automatic tracking of lung nodules using an AI method.…”