Department of Radiation Oncology, University of California San Diego, La Jolla, CA 92037, USA.
Phys Med Biol. 2009 Nov 7;54(21):6565-73. doi: 10.1088/0031-9155/54/21/008. Epub 2009 Oct 14.
The widespread adoption of on-board volumetric imaging in cancer radiotherapy has stimulated research efforts to develop online adaptive radiotherapy techniques to handle the inter-fraction variation of the patient's geometry. Such efforts face major technical challenges to perform treatment planning in real time. To overcome this challenge, we are developing a supercomputing online re-planning environment (SCORE) at the University of California, San Diego (UCSD). As part of the SCORE project, this paper presents our work on the implementation of an intensity-modulated radiation therapy (IMRT) optimization algorithm on graphics processing units (GPUs). We adopt a penalty-based quadratic optimization model, which is solved by using a gradient projection method with Armijo's line search rule. Our optimization algorithm has been implemented in CUDA for parallel GPU computing as well as in C for serial CPU computing for comparison purpose. A prostate IMRT case with various beamlet and voxel sizes was used to evaluate our implementation. On an NVIDIA Tesla C1060 GPU card, we have achieved speedup factors of 20-40 without losing accuracy, compared to the results from an Intel Xeon 2.27 GHz CPU. For a specific nine-field prostate IMRT case with 5 x 5 mm(2) beamlet size and 2.5 x 2.5 x 2.5 mm(3) voxel size, our GPU implementation takes only 2.8 s to generate an optimal IMRT plan. Our work has therefore solved a major problem in developing online re-planning technologies for adaptive radiotherapy.
车载容积成像在癌症放射治疗中的广泛应用刺激了开发在线自适应放射治疗技术的研究工作，以处理患者几何形状的分次间变化。这些努力在实时进行治疗计划方面面临着重大的技术挑战。为了克服这一挑战，我们正在加州大学圣地亚哥分校（UCSD）开发一个超级计算在线重新规划环境（SCORE）。作为 SCORE 项目的一部分，本文介绍了我们在图形处理单元（GPU）上实现调强放射治疗（IMRT）优化算法的工作。我们采用基于惩罚的二次优化模型，该模型通过使用具有 Armijo 线搜索规则的梯度投影法求解。我们的优化算法已经在 CUDA 中实现，用于并行 GPU 计算，并在 C 中实现，用于串行 CPU 计算以进行比较。使用各种射束和体素大小的前列腺 IMRT 病例来评估我们的实现。在 NVIDIA Tesla C1060 GPU 卡上，与 2.27 GHz Intel Xeon CPU 的结果相比，我们实现了 20-40 倍的加速因子，而不会丢失准确性。对于具有 5 x 5 mm(2)射束大小和 2.5 x 2.5 x 2.5 mm(3)体素大小的特定九野前列腺 IMRT 病例，我们的 GPU 实现仅需 2.8 秒即可生成最佳的 IMRT 计划。因此，我们的工作解决了开发自适应放射治疗在线重新规划技术的一个主要问题。
