3D Ultrasound Imaging and Spatial Pressure Measurement System to Investigate Spinal Curve Response Imposed by a Simulated Brace

目前，无法监测脊柱对物理载荷的 3D 机械响应，并且脊柱支撑过程的发展缓慢。本研究旨在研究、推进和集成监测技术（传感器和执行器），特别是实时 3D 超声图像处理技术、3D 无线空间嵌入式压力控制传感器网络和低成本高速 3D 形状采集过程，以研究模拟脊柱支架施加的生物力学载荷如何影响内部脊柱对齐。将完成四项任务：1) 研究用于实时显示的 3D 超声诊断技术，2) 开发原型并在已知几何形状的物理脊柱模型上验证 3D 超声测量，3) 开发和验证空间压力监测系统，以记录施加到身体模型上的压力垫的负载、位置和方向，以及 4) 开发高速自动流程，以使用低成本摄像系统捕获 3D 身体形状。 实时 3D 超声图像将通过低成本便携式超声机和探头内嵌入的位置传感器进行重建和显示，并通过开发的图像处理算法进行增强。将通过应用滤波器、边缘增强和分割来处理超声 3D 数据以重建图像。模型研究将验证 3D 图像开发。 将设计一个定制的测试框架，可容纳多个压力垫，允许 6 度的运动。将开发无线压力控制传感器网络并将其嵌入压力垫中以跟踪空间和压力信息。使用已知负载进行的实验将确定嵌入式压力传感器的精度。 为了开发低成本高速 3D 人体图像采集系统，提出了适用于 Windows 的 Microsoft Kinect v2 传感器。将开发算法来实现高速捕捉，以最大限度地减少身体运动伪影。将进行实验来验证 3D 重建的准确性。我们将开发软件，将 Kinect 中的 3D 点云数据转换为工业雕刻机可以使用的标准 STL 文件格式。 这项研究的结果将是一套工具，可以在没有电离辐射的情况下实时对内部脊柱排列进行成像。 3D 压力跟踪系统可以提供负载和方向信息，并进一步改进用于其他可穿戴计算机应用。高速 3D 形状图像捕获过程可以自动生成可用于其他工业制造应用的表面文件。最终目标意义重大，因为这种新颖的技术包可以提供实时工具来了解脊柱对外力的生物力学反应，并通过自动化加快脊柱支架的设计过程。

Currently, the 3D mechanical response of the spine to physical loadings cannot be monitored and the development of the spinal brace process is slow. This research aims to investigate, advance and integrate monitoring technologies (sensors and actuators), specifically real-time 3D ultrasound image processing techniques, a 3D wireless spatial embedded pressure control sensor network, and a low cost high speed 3D shape acquisition process to investigate how the biomechanical loadings applied from a simulated spinal brace affects the internal spinal alignment. Four tasks will be completed: 1) investigate 3D ultrasound diagnostic technologies for real-time display, 2) develop prototypes and validate 3D ultrasound measurements on a physical spinal phantom with known geometry, 3) develop and validate a spatial pressure monitoring system to record loads, positions and directions of pressure pads applied to a body phantom, and 4) development of a high speed automatic process to capture the 3D body shape using a low cost camera system. Real-time 3D ultrasound images will be reconstructed and displayed with a low cost portable ultrasound machine and embedded position sensor within its probe, enhanced with developed image processing algorithms. Ultrasound 3D data will be processed by applying filters, edge enhancement and segmentation to reconstruct the image. Phantom studies will validate the 3D image development. A customized testing frame accommodating multiple pressure pads which allow 6-degrees of motion will be designed. A wireless pressure control sensor network will be developed and embedded into the pressure pads to track spatial and pressure information. Experiments with known loads will determine accuracy of the embedded pressure sensor. To develop a low cost high speed 3D body image acquisition system, the Microsoft Kinect v2 sensors for Windows are proposed. Algorithms will be developed to attain high speed capture to minimize body motion artifact. Experiments will be performed to validate the accuracy of the 3D reconstruction. Software will be developed to convert the 3D-point cloud data from the Kinect to a standard STL file format that can be used by industrial carving machines. Results of this research will be a suite of tools which can image the internal spinal alignment without ionizing radiation in real-time. The 3D pressure tracker system can provide loading and directional information and be further advanced for other wearable computer applications. The high speed 3D shape image capture process can generate surface files automatically that may be used for other industrial manufacturing applications. The ultimate goal is significant because this novel technology bundle can provide real-time tools to understand the biomechanical response of the spine to external forces and with automation, speed up the spinal brace design process.