Artificial Intelligence Driven Automatic Treatment Planning of Stereotactic Radiosurgery for the Management of Multiple Brain Metastases

人工智能驱动的立体定向放射外科治疗多发性脑转移瘤自动治疗计划

• 批准号: 10501864
• 负责人: Zhen Tian
• 金额: $ 36.59万
• 依托单位: UNIVERSITY OF CHICAGO
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-01 至 2027-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10501864
• 关键词: 3-Dimensional; Affect; Artificial Intelligence; Arts; Beds; Brain; Cancer Patient; Caring; Clinical; Clinical Research; Clinical Trials; Cognition; Communication; Computer Vision Systems; Consultations; Custom; Decision Making; Development; Devices; Disease; Dose; Effectiveness; Evaluation; Excision; Feedback; Foundations; Gamma Knife Radiosurgery; Guidelines; Hour; Human; Intelligence; Learning; Life; Location; Malignant Neoplasms; Manuals; Mathematics; Medical Device; Medicine; Metastatic malignant neoplasm to brain; Methods; Modernization; National Comprehensive Cancer Network; Neurocognition; Neurocognitive Deficit; Normal tissue morphology; Operative Surgical Procedures; Organ; Outcome; Pancreas; Patient Care; Patients; Physicians; Policies; Process; Prospective Studies; Prostate; Psychological reinforcement; Radiation therapy; Radiosurgery; Reporting; Resource-limited setting; Rewards; Robotics; Safety; System; Systems Integration; Technology; Testing; Time; Treatment outcome; Validation; Variant; base; cancer radiation therapy; deep learning; deep neural network; deep reinforcement learning; design; improved; individual patient; innovation; knowledge base; palliative; patient prognosis; preference; prospective; prototype; satisfaction; skills; standard of care; technology development; treatment planning; tumor

Project Summary/Abstract Brain metastases (BMs) are a life-threatening disease, occurring in up to 40% of cancer patients. About 40% of BM patients have multiple (≥4) BMs (mBMs). Whole brain radiotherapy (WBRT), which has long been the standard of care for mBMs patients, has shown pronounced impairment of neurocognitive functions. Stereotactic radiosurgery (SRS) has improved tumor control and reduced negative effect on cognition function, compared to WBRT. However, it has been historically reserved only for patients with <4 BMs. Recently, several clinical trials reported strong evidence to support SRS for mBMs patients. National Comprehensive Cancer Network guidelines hence no longer restrict the number of BMs for SRS. However, the larger BM number in mBMs patients substantially increases the complexity of treatment planning. Conventional manual forward planning to manually determine plan parameters becomes cumbersome and impractical for mBMs. Modern inverse planning methods can determine plan parameters by solving an optimization problem that is composed of multiple objectives designed for various clinical or practical considerations, while the priorities among these objectives affect the resulting plan quality. The physician’s preferences for a particular patient can hardly be quantified and precisely conveyed to the planner, especially for mBMs patients due to the varying number, size, and locations of BMs. Hence, the best physician-preferred plan is often achieved through extensive trial-and-error priority tuning and several rounds of interactions between the planner and physician. Consequently, planning time can take up to hours, and plan quality may be suboptimal and can vary significantly, due to the varying levels of physician and planner’s skills and physician-planner communication and cooperation, leading to deteriorated clinical outcome. Inspired by the recent colossal advancements of artificial intelligence (AI), particularly deep reinforcement learning (DRL) and deep inverse reinforcement learning (DIRL), on intelligent decision-making in computer visions and robotics, we propose to develop an artificial intelligence driven automatic SRS treatment planning system for effective management of mBMs (Aid-mBMs), learning a human-level intelligence on treatment planning from human experts. We envision the system to have two deep neural networks: DNN-R that acts as an AI-physician to predict the physician’s preferences for each individual patient, and DNN-P that acts as an AI-planner to tune the priorities to achieve a plan of physician’s satisfaction. We will pursue two specific aims. Aim 1. System prototype development: We will collect human expert planners’ priority-tuning actions and develop DNN-R and DNN-P via interleaved DIRL-based reward function learning and DRL-based policy learning. Aim 2. System improvement and end-to-end evaluation: We will perform a prospective study to improve our system based on human expert’s further fine-tuning actions on the generated AI plans, and then evaluate the feasibility, effectiveness, and efficiency of our system. Upon completion, Aid-mBMs will provide high-quality and efficient SRS treatment planning to benefit mBMs patients, especially those in resource-limited regions.

项目概要/摘要 脑转移 (BM) 是一种危及生命的疾病，高达 40% 的癌症患者会发生这种情况。 BM 患者接受多次（≥4）次 BM（mBM）治疗，这一直是治疗的重点。 mBM 患者的护理标准已显示出神经认知功能的明显损害。 与传统疗法相比，放射外科 (SRS) 改善了肿瘤控制并减少了对认知功能的负面影响 然而，WBRT 历来仅适用于 BM < 4 的患者，最近进行了多项临床试验。 报告了强有力的证据支持 mBM 患者的 SRS 治疗国家综合癌症网络。 因此，指南不再限制 SRS 的 BM 数量，但是 mBM 中的 BM 数量较多。 患者的传统手动前瞻性计划大大增加了复杂性。 对于现代逆向规划来说，手动确定计划参数变得麻烦且不切实际。 方法可以通过解决由多个组成的优化问题来确定计划参数 为各种临床或实际考虑而设计的目标，而这些目标之间的优先级 影响最终计划的质量。医生对特定患者的偏好很难量化和确定。 准确地传达给计划者，特别是对于 mBM 患者，因为其数量、大小和位置各不相同 因此，最佳的医生首选计划通常是通过广泛的试错优先权来实现的。 计划者和检查医生之间的调整和几轮互动，计划时间可以。 需要长达数小时的时间，并且计划质量可能不是最佳的，并且由于不同水平的不同而可能存在很大差异 医生和规划师的技能以及医生与规划师的沟通与合作，导致恶化 受到人工智能 (AI) 最近巨大进步的启发，特别是深度进步。 强化学习（DRL）和深度逆强化学习（DIRL），关于智能决策 计算机视觉和机器人技术，我们建议开发一种人工智能驱动的自动SRS治疗 有效管理 mBM (Aid-mBM) 的规划系统，学习人类水平的智能 我们设想该系统有两个深度神经网络：DNN-R 充当 AI 医生来预测医生对每个患者的偏好，而 DNN-P 则充当 作为人工智能规划者，我们将调整优先级以实现医生满意的计划。 目标1.系统原型开发：我们将收集人类专家规划者的优先级调整行动和 通过交错的基于 DIRL 的奖励函数学习和基于 DRL 的策略学习来开发 DNN-R 和 DNN-P。 目标2.系统改进和端到端评估：我们将进行前瞻性研究以改进我们的系统 系统基于人类专家对生成的人工智能计划进一步微调动作，然后评估 完成后，Aid-mBM 将提供高质量和高效率的服务。 有效的 SRS 治疗计划可以使 mBM 患者受益，特别是那些资源有限地区的患者。