肝癌放射治疗中自动计划技术的普适性验证

许六军; 李婷婷; 张玉洁; 肖寒; 嵇卫星; 郭明; 曾昭冲; 吴志峰; 张建英

doi:10.12025/j.issn.1008-6358.2021.20202589

肝癌放射治疗中自动计划技术的普适性验证

Feasibility validation of combined automated planning technique for liver cancer in radiotherapy

摘要

摘要:
目的探讨放射治疗中基于知识的肝癌自动计划应用于4种Varian模型的可行性。
方法将基于知识的自动计划约束条件应用4种加速器模型，4种模型涉及不同的机器类型、能量、叶片对数和调强技术，分别在每种模型上应用20例患者，并分别完成手动和自动计划，通过靶区和危及器官的剂量比较评估计划质量。
结果 Step-and-shoot技术结合自动计划能得到更优的计划（P < 0.05）。在所有4种机器模型中，手动计划和自动计划的靶区适形性从0.83~0.84提高到0.86~0.87，而正常肝的平均剂量和左肾平均剂量分别从1 587.5~1 782.5 cGy、299.8~362.3 cGy降低至1 480.1~1 669.0 cGy、138.3~209.4 cGy，正常肝的V₁₅从420.7~493.4 mL降低至383.2~455.2 mL，差异有统计学意义（P < 0.05）。
结论基于知识的自动计划在多种机器模型中均有效。

Abstract:
Objective To explore the feasibility of applying a knowledge-based protocol on four Varian models with Auto-Planning for liver cancer.
Methods Our previously developed knowledge-based protocol for Auto-Planning was applied on four Varian models, in which different machine types, energies, numbers of leaf pairs, and IMRT techniques were involved. Manual and automated plans were accomplished for twenty tested patients on each model. Plan quality was estimated by dose comparison of target and organ at risks (OARs).
Results Step-and-shot technology combined with automatic planning can get better plan (P < 0.05). In all four machine models, the target conformability of manual planning and automatic planning increased from 0.83-0.84 to 0.86-0.87, while the average dose and average dose of left kidney of normal liver decreased from 1 587.5-1 782.5 cGy and 299.8-362.3 cGy to 1 480.1-1 669.0 cGy and 138.3-209.4 cGy, respectively. The V₁₅ decreased from 420.7-493.4 mL to 383.2-455.2 mL, the difference was statistically significant(P < 0.05).
Conclusions Along with Auto-Planning, the knowledge-based protocol is effective in multiple machine systems.

HTML全文

参考文献(24)

施引文献

资源附件(0)