RESUMO
Objective:To provide a basis for selecting the optimization method for intracavitary/interstitial brachytherapy (IC/ISBT) of cervical cancer by comparing graphical optimization (GO), inverse planning simulated annealing (IPSA), and hybrid inverse planning optimization (HIPO) using dosimetric and radiobiological models.Methods:This study selected 65 patients with cervical cancer who were treated with image-guided IC/ISBT. The afterloading therapy plans for these patients were optimized using GO, IPSA, and HIPO individually, with a prescription dose high-risk clinical target volume (HRCTV) D90 of 6 Gy. The non-parametric Friedman test and the non-parametric Wilcoxon rank test were employed to analyze the differences in duration, dose-volume parameters, and radiobiology between the three types of optimized plans. Results:Inverse planning optimization (IPSA: 46.53 s; HIPO: 98.36 s) took less time than GO (135.03 s). In terms of gross target volume (GTV) dose, the high-dose irradiation V150% (53.66%) was slightly higher in the HIPO-optimized plans, while the V200% (30.29%) was higher in the GO-optimized plans. The GO-optimized plans had a higher conformity index (CI; 0.91) than other plans, showing statistically significant differences. Compared with other plans, the HIPO-optimized plans showed the lowest doses of D1 cm 3 and D2 cm 3 at bladders and rectums and non-statistically significant doses at small intestines ( P > 0.05). In terms of the equivalent uniform biologically effective dose (EUBED) for HRCTV, the HIPO-optimized plans showed a higher value (12.35 Gy) than the GO-optimized plans (12.23 Gy) and the IPSA-optimized plans (12.13 Gy). Moreover, the EUBED at bladders was the lowest (2.38 Gy) in the GO-optimized plans, the EUBED at rectums was the lowest (3.74 Gy) in the HIPO-optimized plans, and the EUBED at small intestines was non-significantly different among the three types of optimized plans ( P = 0.055). There was no significant difference in the tumor control probability (TCP) predicted using the three types of optimized plans ( P > 0.05). The normal tissue complication probabilities (NTCPs) of bladders and rectums predicted using the HIPO-optimized plans were lower than those predicted using the GO- and IPSA-optimized plans( χ2 = 12.95-38.43, P < 0.01), and the NTCP of small intestines did not show significant differences ( P > 0.05). Conclusions:Among the three types of optimization algorithms, inverse optimization takes less time than GO. GO-optimized plans are more conformal than IPSA- and HIPO-optimized plans. HIPO-optimized plans can increase the biological coverage dose of the target volume and reduce the maximum physical/biological exposure and NTCP at bladders and rectums. Therefore, HIPO is recommended preferentially as an optimization algorithm for IC/ISBT for cervical cancer.
RESUMO
Objective:To establish a three-dimensional (3D) U-net-based deep learning model, and to predict the 3D dose distribution in CT-guided cervical cancer brachytherapy by using the established model.Methods:The brachytherapy plans of 114 cervical cancer cases with a prescription dose of 6 Gy for each case were studied. These cases were divided into training, validation, and testing groups, including 84, 11, and 19 patients, respectively. A total of 500 epochs of training were performed by using a 3D U-net model. Then, the dosimetric parameters of the testing groups were individually evaluated, including the mean dose deviation (MDD) and mean absolute dose deviation (MADD) at the voxel level, the Dice similarity coefficient (DSC) of the volumes enclosed by isodose surfaces, the conformal index (CI) of the prescription dose, the D90 and average dose Dmean delivered to high-risk clinical target volumes (HR-CTVs), and the D1 cm 3 and D2 cm 3 delivered to bladders, recta, intestines, and colons, respectively. Results:The overall MDD and MADD of the 3D dose matrix from 19 cases of the testing group were (-0.01 ± 0.03) and (0.04 ± 0.01) Gy, respectively. The CI of the prescription dose was 0.70 ± 0.04. The DSC of 50%-150% prescription dose was 0.89-0.94. The mean deviation of D90 and Dmean to HR-CTVs were 2.22% and -4.30%, respectively. The maximum deviations of the D1 cm 3 and D2 cm 3 to bladders, recta, intestines, and colons were 2.46% and 2.58%, respectively. The 3D U-net deep learning model took 2.5 s on average to predict a patient′s dose. Conclusions:In this study, a 3D U-net-based deep learning model for predicting 3D dose distribution in the treatment of cervical cancer was established, thus laying a foundation for the automatic design of cervical cancer brachytherapy.