Development of Novel Fast GPU Monte Carlo and Active Photonics Simulation Software for Predicting PDT Efficacy

 描述(申请人提供)：光动力疗法(PDT)提供独立或术中癌症治疗。光动力疗法提供了一种治疗浅表和/或残留疾病的方法，同时将对底层组织的损害降至最低，并且不会表现出累积的细胞毒性，这与放射治疗不同。与放射治疗相比，光动力疗法的治疗计划通常是一刀切的。然而，患者和肿瘤特有的因素，如组织光学性质和光敏剂(PS)水平，对有效光剂量的传递至关重要。考虑到这些因素的治疗剂量学工具的开发将填补尚未满足的临床需求，并提供个性化的患者治疗。一个有效的PDT治疗剂量测定系统将改善治疗结果，减少重复PDT或额外的细胞杀伤治疗的需要，因此可以降低每个患者提供与癌症相关的治疗和护理的总体成本。SBIR第二阶段提案的主要目标是开发和验证原型软件和硬件工具，这些工具将PS光物理模拟与使用快速蒙特卡罗(MC)技术的光传播相结合。这一第二阶段SBIR的研究将产生独特的原型剂量学工具，这些工具将在第三阶段进一步开发和商业化，供PDT医生和研究人员使用，以改善患者的预后。这项第二阶段SBIR有三个主要目标。目标1是开发利用快速蒙特卡罗(MC)技术和患者PS可变性相结合的光传输的PDT剂量测量原型软件。该软件应该足够快，以便将来在该项目的第三阶段临床使用。在该系统的基础上，SimPhotek将推出新的主动光子积木(APBB)算法，该算法具有用于主动光物理的简单图形用户界面。APBB将光物理的计算问题分解为一系列计算构件，软件自动组合这些构件来生成完整的数值模拟。为了将光散射纳入分析，SimPhotek与高性能计算领域的领先者Tech-X Corporation(Tech-X；SUBAWARD)进行了合作。Tech-X已经开发了基于MC的散射基础设施，并对第一阶段的代码进行了修改，以模拟生物组织中的光扩散和吸收过程。目标2的目标是让Tech-X和SimPhotek开发一个原型PDT剂量学工具，将目标1中开发的软件和用于高速模拟的专用硬件结合在一起。目标3是通过将模拟结果与由PDT专家在宾夕法尼亚大学医学院(宾夕法尼亚大学；分奖)所做的体模测量进行比较，来验证软件/硬件模拟。

 DESCRIPTION (provided by applicant): Photodynamic therapy (PDT) provides for standalone or intraoperative cancer treatment. PDT provides a means to treat superficial and/or residual disease, while minimizing damage to underlying tissues, and does not exhibit cumulative cell toxicities, distinguishing it from radiation therapy. As compared to radiotherapy, treatment planning in PDT is often approached in a one-size-fits-all fashion. However, patient- and tumor-specific factors such as tissue optical properties and photosensitizer (PS) levels are critical to the delivery of effective light doses. The development of treatment dosimetry tools tha take into account these factors will fill an unmet clinical need and provide for individualized patient treatment. An effective PDT treatment dosimetry system stands to improve therapeutic outcomes, reduce the need for repeat PDT or additional cell-killing therapy, and could therefore reduce overall costs in the per patient delivery of cancer-related therapy and care. The major objective of this SBIR Phase II proposal is to develop and verify prototype software and hardware tools that combine simulations of PS photophysics with light propagation using fast Monte-Carlo (MC) techniques. The research of this Phase II SBIR will result in unique prototype dosimetry tools that will be further developed and commercialized in Phase III for use by PDT physicians and researchers to improve patient outcomes. This Phase II SBIR has three major aims. Aim 1 is to develop prototype software for PDT dosimetry combining light transport using fast Monte-Carlo (MC) techniques and patient PS variability. The software should be fast enough for future clinical use in Phase III of this project. At the foundation of this system will e Simphotek's novel Active Photonics Building Blocks (APBB) algorithm with its simple graphical user interface for active photophysics. The APBB breaks the computing problem for photophysics into a series of computational building blocks that the software automatically combines to generate the full numerical simulation. To include light scattering in the analysis, Simphotek has partnered with Tech-X Corporation (Tech-X; subaward), a leader in the field of high-performance computing. Tech-X has developed MC-based scattering infrastructure and has adapted the code in Phase I to model light diffusion and absorption processes in biological tissue. The Aim 2 objective is for Tech-X and Simphotek to develop a prototype PDT dosimetry tool combining both the software developed in Aim 1 and specialized hardware for high-speed simulations. Aim 3 is to verify the software/hardware simulations by comparing the simulation results to phantom measurements done at the University of Pennsylvania School of Medicine (Penn; subaward) by experts in PDT.