计算机工程与应用 ›› 2020, Vol. 56 ›› Issue (17): 197-202.DOI: 10.3778/j.issn.1002-8331.1906-0380

• 图形图像处理 • 上一篇    下一篇

结合增强现实的医用加速器碰撞检测方法

张功森,郭翌,李永哲,裴曦,徐榭,周解平   

  1. 1.中国科学技术大学 物理学院 工程与应用物理系,合肥 230022
    2.中国科学技术大学附属第一医院西区,合肥 230001
  • 出版日期:2020-09-01 发布日期:2020-08-31

Medical Accelerator Collision Detection Method Combined with Augmented Reality

ZHANG Gongsen, GUO Yi, LI Yongzhe, PEI Xi, XU Xie, ZHOU Jieping   

  1. 1.Department of Engineering and Applied Physic, School of Physical Science, University of Science and Technology of China, Hefei 230022, China
    2.West District, First Affiliated Hospital of University of Science and Technology of China, Hefei 230001, China
  • Online:2020-09-01 Published:2020-08-31

摘要:

针对肿瘤放射治疗中的加速器机械碰撞问题,提出一种增强现实环境下基于Unity3D的碰撞检测方法。基于三维建模软件3ds Max对医用直线加速器、治疗床等设备进行准确建模,基于病人CT由移动立方体算法得到三维可视化图像,并据此在USTC系列人体模型数据库中匹配体型符合程度最高的模型;编写脚本以对模型进行运动控制及刚体特性处理,基于Unity3D进行碰撞检测并规划设计GUI界面以进行用户操作和信息反馈;将三维模拟场景发布到Universal Windows平台并部署到增强现实设备HoloLens上。针对病床(病人)的三组空间位置进行测试,检测结果均与实际情况相符合,并得到了准确的反馈信息。测试结果表明,该碰撞检测方法检测结果准确,人机交互方便,增强现实环境下,展示效果直观,具有很好的应用前景。

关键词: 医用加速器, 碰撞检测, 增强现实, 三维可视化

Abstract:

Aiming to solve the problem of accelerator mechanical collision in tumor radiotherapy, a unity3D-based collision detection method based on augmented reality is proposed. Firstly, all equipments are accurately modeled based on the 3D modeling software 3ds Max. Based on the CT of patient, the 3D visualization image is obtained by marching cubes algorithm, and the model with the highest conformity is matched in the USTC phantom database. Then the scripts is used to control the model and rigid body feature processing, and perform collision detection based on Unity3D and design the GUI interface for user’s operation and information feedback. Finally, the 3D simulation scene is published to the Universal Windows platform and deployed to the augmented reality. The three cases of spatial patient position are tested. The results are consistent with the actual situation, and the accurate feedback information is obtained. The test results show that proposed method is accurate, and the human-computer interaction is convenient and intuitive, so it has a good application prospect.

Key words: medical accelerator, collision detection, augmented reality, 3D visualization