Improving absorbed dose estimation for treatment planning in Molecular Radiotherapy

改进分子放射治疗中治疗计划的吸收剂量估计

• 批准号: 2734835
• 金额: --
• 依托单位: University College London
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2734835
• 关键词: Improving; absorbed; dose; estimation; treatment

Molecular Radiotherapy (MRT) is a rapidly growing cancer treatment modality where molecules that bind to cancerous cells are labelled with a radionuclide and injected into the patient for targeted delivery of radiation. Multimodality imaging using CT and nuclear imaging (SPECT/PET) can be performed to quantify absorbed doses to the tumours and organs at risk. Nevertheless, personalised treatments have not yet made it into routine clinical use. This is partly due to a lack of standardisation and knowledge on the uncertainties in dosimetry calculations as well as increased resources needed, leading to a lack of evidence from large randomised clinical trials.Machine learning techniques are under investigation for nuclear medicine dosimetry but have not yet been implemented clinically due to the lack of validation and knowledge on their potential benefits. In particular, deep learning models have been proposed as a way to increase the speed of the absorbed dose calculation step as well as decreasing the need for imaging resources.The aim is to contribute towards the implementation of personalised treatment planning into clinical practice by characterising and improving accuracy and uncertainty of dosimetry calculations from SPECT and CT images. The student will compare conventional methods for absorbed dose map generation with state-of-the-art Machine Learning methods. Sensitivity to acquisition protocol and reconstruction method will be investigated as well, with a view to optimise and simplify protocols (e.g. towards single time point dosimetry).The project will include the development and validation of a simulation framework for generating a realistic ground truth dataset using open source software GATE/ STIR / OpenDose / Dositest, based on existing clinical data from theragnostic studies at UCLH. Some experiments using phantoms will have to be performed at UCLH and NPL.In line with EPSRC strategy and research areas the project will assist in the creation of safer and more targeted patient specific radiation treatments. Novel imaging technologies will be explored including different imaging reconstruction techniques in multi-modal imaging systems and Machine Learning methods. Optimisation and simplification of protocols will accelerate translation of novel radiation therapies in real world clinical applications.This project is a collaboration between the Institute of Nuclear Medicine (INM), and the nuclear medicine group of the National Physical Laboratory. The student will be part of the UCL i4health Centre for Doctoral Training (CDT) and the postgraduate institute (PGI) at NPL, and will benefit from a wide range of activities and opportunities. The student will be primarily located at INM, near the UCL Bloomsbury Campus. Imaging facilities include SPECT-CT, PET-CT and PET-MRI scanners. A substantial proportion of the student's time will be spent at the NPL Teddington Campus for collaboration and experiments. The supervisory team will include Prof Kris Thielemans (UCL), Dr Sarah McQuaid (UCLH) and Dr Ana Denis-Bacelar (NPL).

分子放射治疗（MRT）是一种快速发展的癌症治疗方式，将与癌细胞结合的分子标记为放射性核素，并将其注射到患者体内，以进行靶向放射治疗。可以使用CT和核成像（SPECT/PET）进行多模态成像，以量化肿瘤和危险器官的吸收剂量。然而，个性化治疗尚未进入常规临床应用。这在一定程度上是由于缺乏对剂量学计算不确定性的标准化和认识，以及所需资源的增加，导致缺乏来自大型随机临床试验的证据。机器学习技术正在研究用于核医学剂量测定，但由于缺乏对其潜在益处的验证和知识，尚未在临床上实施。特别是，深度学习模型已被提出作为提高吸收剂量计算步骤速度以及减少对成像资源需求的一种方法。目的是通过表征和提高SPECT和CT图像剂量学计算的准确性和不确定性，为临床实践中个性化治疗计划的实施做出贡献。学生将比较传统的吸收剂量图生成方法与最先进的机器学习方法。还将研究对采集方案和重建方法的敏感性，以优化和简化方案（例如，单时间点剂量测定）。该项目将包括开发和验证一个模拟框架，用于基于UCLH治疗研究的现有临床数据，使用开源软件GATE/ STIR / OpenDose / Dositest生成现实的地面真实数据集。一些使用幻影的实验将不得不在UCLH和NPL进行。根据EPSRC的战略和研究领域，该项目将协助创建更安全、更有针对性的患者特异性放射治疗。将探索新的成像技术，包括多模态成像系统中的不同成像重建技术和机器学习方法。协议的优化和简化将加速新的放射治疗在现实世界的临床应用的翻译。该项目由核医学研究所（INM）和国家物理实验室核医学小组合作完成。该学生将成为UCL i4博士培训健康中心（CDT）和NPL研究生研究所（PGI）的一部分，并将从广泛的活动和机会中受益。该学生将主要位于伦敦大学学院布卢姆斯伯里校区附近的INM。成像设备包括SPECT-CT， PET-CT和PET-MRI扫描仪。学生的大部分时间将在国家物理实验室特丁顿校区进行合作和实验。监督小组将包括克里斯·蒂勒曼斯教授（伦敦大学学院）、莎拉·麦奎德博士（伦敦大学学院）和安娜·丹尼斯-巴塞拉博士（国家物理实验室）。