Author: Zihao Liu, Qiwei Wu, Yanfei Xiong, Yidong Yang, Ning Zhao 👨🔬
Affiliation: Department of Engineering and Applied Physics, School of Physical Sciences, University of Science and Technology of China, Department of Engineering and Applied Physics, University of Science and Technology of China, University of Science and Technology of China 🌍
Purpose: To develop an inverse planning framework that optimizes beam angles and intensities for small animal radiotherapy and to validate its accuracy and effectiveness.
Methods: The inverse planning framework began with generating the dose matrix for 144 candidate X-ray beams. The radiation dose from each beam was calculated using a GPU-based Monte-Carlo dose calculation engine. The objective function was designed to minimize the quadratic differences between the prescribed dose and planned dose. Beam angle optimization was solved through a fast iterative hard thresholding algorithm. Four lung tumor cases were employed to compare the performances of forward manual plans and inverse optimized plans. To validate the dosimetric accuracy, the plan dose was compared against the measured delivered doses. Furthermore, animal experiments using mouse lung cancer models were carried out to further evaluate the feasibility and effectiveness of the proposed framework in small animal radiotherapy.
Results: Both manual and inverse treatment plans met the prescribed doses, with no significant differences in target coverage and uniformity. However, in inverse-optimized treatment plans, the maximum dose to the spinal cord decreased by 54.10% ± 6.43% (p = 0.001), and the mean dose to the heart decreased by 22.97% ± 8.94% (p = 0.020). In phantom radiation experiments, under the irradiation of five beams with 5 mm and 7.5 mm diameter collimators, the differences between calculated and measured doses were less than 5% and the gamma passing rate was 91.7% and 89.8% at 3%/0.3 mm criteria, respectively. Histological analysis of irradiated lungs confirmed that the inverse plans can be accurately delivered to tumor targets in live animals.
Conclusion: The proposed inverse planning framework effectively reduces the dose to critical organs while maintaining target coverage and uniformity. The dose delivery accuracy was validated in both phantom and animal experiments.