Distributed knowledge-based platform for radiotherapy plan quality control

分布式放疗计划质量控制知识平台

• 批准号: 9900745
• 负责人: Kevin Lawrence Moore
• 金额: $ 37.46万
• 依托单位: UNIVERSITY OF CALIFORNIA, SAN DIEGO
• 依托单位国家: 美国
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-06-11 至 2023-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9900745
• 关键词: Distributed; knowledge; based; platform; radiotherapy

ABSTRACT Many recent studies focused on radiotherapy treatment plan quality have begun to quantify what clinicians have long understood: even “optimized” radiotherapy is no guarantee of a truly optimal treatment plan for every patient. Plan quality deficiencies have been shown to put a significant proportion of patients who should have been at low risk of radiation-induced complications at much higher risk for poor outcome. Available research clearly demonstrates a link between radiation provider volume and survival, which emphasizes the importance of quality radiation delivery. Radiation providers in rural or community practices by nature see a wide variety of cases, with lower provider volume for each individual disease site. Through no fault of their own, physician and non-physician practitioners at these rural and community centers could be inadvertently and systematically delivering low quality radiotherapy to their patients simply due to the fact that no platform currently exists that could benchmark their practice against a distributed, externally-validated plan quality control system. Our research team has developed, tested, and clinically-implemented an important tool to combat radiotherapy plan quality deficiencies known as knowledge-based planning (KBP). Knowledge-based planning relies on the use of statistical learning techniques that analyzed a plurality of prior treatments to discover patient-specific anatomical features can be precisely correlated to high quality radiation dose delivery. Unfortunately, the clinical use of KBP has been limited to a handful of high-volume academic centers and, without some external mechanism to increase utilization, its use is not likely to expand significantly to rural and community centers because of the lack of any billing code associated with its use. To provide just such an external mechanism, we intend to build ORBITeR (On-line Real-time Benchmarking Informatics Technology for Radiotherapy), a freely available, on-line knowledge-based radiotherapy plan quality control system. ORBITeR will allow clinicians to obtain automatic and immediate feedback on the quality of any individual treatment plan prior to treatment. We will develop a KBP-driven plan analysis system on a HIPAA-compliant web-based platform designed to give users real- time radiotherapy plan quality feedback. To provide real-time feedback to clinical users, we will develop reporting modules on the ORBITeR system that provide patient-specific feedback on the quality of the intended treatment plan using already-validated head-and-neck, brain, prostate, cervix, lung, pancreas, and liver cancer knowledge-based models. We then will disseminate and evaluate the effectiveness of the ORBITeR plan quality resource among the greater radiation oncology community. Finally, we will develop a quality analytics system to conduct widespread plan quality and patterns of care study across submitting sites on the ORBITeR system. .

摘要 许多最近的研究集中在放射治疗计划的质量已经开始量化什么 临床医生早就明白：即使是“优化”的放射治疗也不能保证真正的最佳治疗效果。 为每位患者制定治疗方案。计划质量缺陷已被证明是一个重大的 本应处于辐射引起的并发症低风险的患者比例 结果不佳的风险更高。现有的研究清楚地表明，辐射提供者之间的联系 体积和生存率，强调了高质量放射治疗的重要性。辐射提供者 在农村或社区的做法，从性质上看，看到各种各样的情况下，与较低的供应商量， 每一个单独的疾病部位。通过没有自己的过错，医生和非医生从业者在 这些农村和社区中心可能会无意中系统地提供低质量的服务， 因为目前还没有一个平台可以 将他们的实践与分布式的、外部验证的计划质量控制系统进行基准测试。我们 一个研究小组已经开发，测试和临床实施了一个重要的工具，以打击 放疗计划质量缺陷称为基于知识的计划（KBP）。知识型 规划依赖于使用统计学习技术， 发现患者特异性解剖特征可以与高质量辐射剂量精确相关 交付.不幸的是，KBP的临床应用仅限于少数高容量的学术研究。 如果没有一些外部机制来提高利用率，其使用不太可能扩大 由于缺乏与其相关的任何计费代码， 使用.为了提供这样一个外部机制，我们打算建立ORBITeR（在线实时 基准信息技术放射治疗），一个免费提供的，在线的知识为基础的 放射治疗计划质量控制系统。ORBITeR将允许临床医生获得自动和 在治疗前对任何个人治疗计划的质量进行即时反馈。我们将开发一个 KBP驱动的计划分析系统，基于符合HIPAA的网络平台，旨在为用户提供真实的 及时反馈放疗计划质量。为了向临床用户提供实时反馈，我们将开发 ORBITeR系统上的报告模块，提供患者对治疗质量的特定反馈 预期治疗计划使用已确认的头颈部、脑、前列腺、宫颈、肺、胰腺， 和肝癌知识模型。然后，我们将传播和评估 ORBITeR计划在更大的放射肿瘤学社区中提供优质资源。最后我们将 开发一个质量分析系统，以进行广泛的计划质量和护理模式研究， 在ORBITeR系统上提交网站。 .