Department of Information and Communication, Flensburg University of Applied Sciences, Kanzleistraße 91-93, 24943, Flensburg, Germany.
Department of Information and Communication, University of Applied Sciences and Arts Hannover, Ricklinger Stadtweg 120, 30459, Hannover, Germany.
Int J Comput Assist Radiol Surg. 2020 Apr;15(4):691-702. doi: 10.1007/s11548-020-02126-x. Epub 2020 Mar 4.
Scattered radiation, which occurs when using a C-arm for intraoperative radiography, can be better understood through interactive visualization. We developed a virtual reality (VR) approach for the simulation of scattered radiation (SSR) as part of a C-arm training system. In VR, it is important to avoid cyber sickness, which is often caused by increased latency between head motion and image presentation inside the head-mounted display. As the latency requirement interferes with the computational complexity of the SSR, the goal has been to maintain a low latency during the simultaneous computation of the SSR on moderate-cost consumer hardware.
For use with a VR C-arm simulator, a CUDA-based Monte Carlo SSR has been improved to utilize GPU resources unused by the VR image generation. Resulting SSR data are visualized through volume rendering with pseudo-colored scattered radiation superimposed onto the virtual operating room. The resulting interactive VR-SSR environment was evaluated with operating room personnel (ORP) and surgeons using questionnaires.
Depending on the imaged body part and computation parameters, the required computation time to complete one SSR run was between 1.6 and 4.2 s (ankle) and between 7.9 and 14.9 s (thigh), and VR frame times from 11 to 12 ms (95th percentile). The system was evaluated with ORP (n = 46) and surgeons (n = 25). The median of professional C-arm experience was 5 (range 1 to 34) years (ORP) and 12.5 (range 2 to 48) years (surgeons), respectively. The demonstrated prototype was found useful by 78% of ORP and 88% of the surgeons. On a Likert scale, more than 90% of both groups "agreed fully" that the presented way of visualizing SSR in VR helps understanding intraoperative exposure to scattered radiation.
Leveraging off-the-shelf computer equipment, the feasibility of SSR and VR for interactive training has been demonstrated. Evaluation participants showed a high interest for the presented approach. Feedback suggests that the visualization experienced by the users helps understanding radiation hazards in the operating room.
术中使用 C 臂进行放射摄影时会产生散射辐射,通过交互式可视化可以更好地理解散射辐射。我们开发了一种虚拟现实(VR)方法来模拟散射辐射(SSR),作为 C 臂培训系统的一部分。在 VR 中,避免网络眩晕症很重要,网络眩晕症通常是由于头部运动和头部显示器内图像呈现之间的延迟增加引起的。由于延迟要求会干扰 SSR 的计算复杂性,因此目标是在中等成本的消费类硬件上同时计算 SSR 时保持低延迟。
为了与 VR C 臂模拟器一起使用,我们改进了基于 CUDA 的蒙特卡罗 SSR,以利用 VR 图像生成未使用的 GPU 资源。通过体绘制将叠加在虚拟手术室中的伪彩色散射辐射的 SSR 数据可视化。使用问卷调查对手术室人员(ORP)和外科医生对产生的交互式 VR-SSR 环境进行了评估。
根据成像的身体部位和计算参数,完成一次 SSR 运行所需的计算时间在 1.6 到 4.2 秒(脚踝)和 7.9 到 14.9 秒(大腿)之间,VR 帧时间为 11 到 12 毫秒(95 百分位)。该系统已由 ORP(n=46)和外科医生(n=25)进行了评估。ORP 的专业 C 臂经验中位数为 5 年(范围为 1 至 34 年),外科医生为 12.5 年(范围为 2 至 48 年)。78%的 ORP 和 88%的外科医生认为展示的原型有用。在李克特量表上,两组中有超过 90%的人“完全同意”,表示在 VR 中以呈现的方式可视化 SSR 有助于理解术中暴露于散射辐射。
利用现成的计算机设备,已经证明了 SSR 和 VR 进行交互式培训的可行性。评估参与者对所提出的方法表现出了浓厚的兴趣。反馈表明,用户体验的可视化有助于理解手术室中的辐射危害。