AcceleraDesign and optimisation of ultra-compact, high-resolution X-ray imaging systems

蒙特卡罗方法已显示出其潜力的真实再现医学X射线图像使用人体幻影。在这个项目中，我们正在利用这种潜力来研究一种新型X射线成像系统的详细行为。该系统是由Adaptix Ltd开发的3D成像系统，具有45个单独的X射线源，排列在矩形网格中。发射器的位置提供了45个不同的辐射投影角度，即45个不同的图像，然后可以通过数学方法进行组合，以创建对象的3D重建，这种方法称为数字断层合成。辐射输出，剂量和所产生的图像质量都使用Monte Carlo代码FLUKA和Geant 4进行研究。然后，系统在减少输送到患者和环境的剂量、改善图像质量和提高X射线源设计效率方面进行优化。这种端到端的仿真框架允许优化集中在系统的特定组件或更一般的水平上，从而提供定制和整体改进能力。此外，两种蒙特卡罗代码都用于实现各种类型的成像体模，包括逼真的人体体模，以模拟实验室中常规进行的质量测试。因此，模拟结果可以与实验数据以及彼此进行基准测试。一旦验证，模型可以用来模拟一系列的实验条件，并执行虚拟测量，至少是非常困难的，如果不是不可能的，进行experimental.Finally，蒙特卡罗方法被用来研究设计概念，以减少系统的足迹，使其更轻，针对一个紧凑和移动的X射线成像系统。这项工作有助于减少资源和实验成本，从而有助于整体降低产品成本。所有工作都是与Adaptix密切合作完成的，Adaptix还实现了延长借调。

Monte Carlo methods have shown their potential for the realistic reproduction of medical X-ray images using human phantoms. In this project, we are using this potential to study the detailed behaviour of a novel X-ray imaging system. The system is a 3D imaging system developed by Adaptix Ltd and has 45 individual X-ray sources, arranged in a rectangular grid. The positions of the emitters offer 45 different projection angles of the radiation, i.e. 45 different images which can then be combined mathematically to create a 3D reconstruction of the object, a method called digital tomosynthesis.The radiation output, dose and resulting image quality are all investigated using the Monte Carlo codes FLUKA and Geant4. The system is then optimised with regards to reducing the dose delivered to the patient and environment, improving image quality and increasing the efficiency of the X-ray source design. This end-to-end simulation framework allows the optimisation to be focused either on a specific component of the system or on a more general level, thus offering both, tailored and overall improvement capabilities.In addition, both Monte Carlo codes are used to implement various types of imaging phantoms, including realistic human phantoms, to simulate quality tests that are routinely done in the laboratory. As a result, the simulation results can be benchmarked against experimental data and against each other. Once validated, the models can be used to simulate a range of experimental conditions and perform virtual measurements that are at least very difficult, if not impossible, to conduct experimentally.Finally, Monte Carlo methods are used to investigate designs concept to reduce the footprint of the system and make it lighter, targeting a compact and mobile X-ray imaging system. This work helps reduce resources and experimental costs and thus contributes to an overall reduction in product cost. All work is realized in close collaboration with Adaptix, where an extended secondment was also realized.