Kang Ki Mun, Choi Hoon Sik, Jeong Bae Kwon, Song Jin Ho, Ha In-Bong, Lee Yun Hee, Kim Chul Hang, Jeong Hojin
Department of Radiation Oncology, Gyeongsang National University School of Medicine and Gyeongsang National University Hospital, Jinju, Republic of Korea.
Institute of Health Sciences, Gyeongsang National University, Jinju, Republic of Korea.
Oncotarget. 2017 May 7;8(33):54497-54505. doi: 10.18632/oncotarget.17672. eCollection 2017 Aug 15.
An alternative pseudo CT generation method for magnetic resonance image (MRI)-based radiotherapy planning was investigated in the work. A pseudo CT was initially generated using the rigid image registration between the planning MRI and previously acquired diagnostic CT scan. The pseudo CT generated was then refined to have the same morphology with that of the referenced planning image scan by applying the outer body correction scheme. This method was applied to some sample of brain image data and the feasibility of the method was assessed by comparing dosimetry results with those from the current gold standard CT-based calculations. Validation showed that nearly the entire pixel doses calculated from pseudo CT were agreed well with those from actual planning CT within 2% in dosimetric and 1mm in geometric uncertainty ranges. The results demonstrated that the method suggested in the study was sufficiently accurate, and thus could be applicable to MRI-based brain radiotherapy planning.
在这项工作中,研究了一种用于基于磁共振成像(MRI)的放射治疗计划的替代伪CT生成方法。首先,通过规划MRI与先前获取的诊断CT扫描之间的刚性图像配准生成伪CT。然后,通过应用体外校正方案对生成的伪CT进行细化,使其形态与参考规划图像扫描相同。该方法应用于一些脑图像数据样本,并通过将剂量学结果与当前基于CT的金标准计算结果进行比较来评估该方法的可行性。验证表明,在剂量学不确定度范围为2%、几何不确定度范围为1mm内,由伪CT计算得到的几乎所有像素剂量与实际规划CT计算得到的剂量吻合良好。结果表明,该研究中提出的方法足够准确,因此可应用于基于MRI的脑部放射治疗计划。