Towards New Frontiers in Computer Assisted Image Guidance with Next Generation Point of Care Ultrasound

近年来，使用实时超声成像的图像引导解决方案彻底改变了计算机辅助干预系统。然而，目前的系统仅限于配备昂贵的专业运动跟踪设备和先进、昂贵的成像设备的专业中心。这些系统大规模分布的一个解决方案是护理点超声，它使移动设备上的图像采集成本低得多。然而，潜在的护理点操作员是新手/未经超声培训。为了获得正确的超声成像数据，直观地解释数据，并利用超声作为有效的指导，操作人员需要手眼协调。我们需要的是技术上的重大飞跃，以提高超声在特定图像引导应用中的易用性。***该研究项目将开发基础技术，以引领下一代计算机辅助干预系统的转型，使其具备即时超声设备。通过开发新的方法，使用自适应波束形成准确地确定潜在的组织特性，可以实现更好的数据采集。组织特性的准确估计将用于增强3D超声体积的重建，而不需要体积庞大且阻碍工作流程的专门跟踪设备。***通过使用机器学习解决方案，可以自动识别3D超声中的关键解剖标志，分割相关解剖边界，并将3D超声与术前图像融合以利用互补的解剖信息，从而解决更好的解释问题。验证将在迭代设计过程中建立。设计应用的一个例子是脊髓麻醉，用于指导硬膜外针注射程序。***所有这些技术将单独或一起在商业点护理超声机上实施，从加拿大制造的Clarius开始。这些技术将使包括脊髓麻醉在内的一些图像引导手术比目前最先进的手术更容易、更准确。因此，这项研究计划将使图像引导技术的广泛部署成为可能。***拟培养博士3人，硕士3人，本科生5人。一个详细的培训计划，使个人能够充分发挥其潜力，并与研究目标的发展相结合，以确保HQP根据UBC公平和多样性战略计划和政策#2的高质量，互动，繁荣和公平的研究环境。

In recent years, image-guided solutions using real-time ultrasound imaging have revolutionized computer assisted intervention systems. However, current systems are limited to specialized centres equipped with expensive specialized motion tracking equipment and advanced, expensive imaging devices. A solution to large-scale distribution of these systems is point-of-care ultrasound, which enables image acquisition on mobile devices at much lower cost. However, potential point-of-care operators are novices/untrained in ultrasound. Operators struggle with the hand-eye coordination needed to obtain the correct ultrasound imaging data, intuitively interpret the data, and to use ultrasound as an effective guide. What is needed is a significant leap in technologies that will improve the ease-of-use of ultrasound for specific image-guided applications.***This research program will develop fundamental technologies to lead the transformation of next generation computer-assisted intervention systems, enabled with point-of-care ultrasound devices. Better data acquisition will be achieved by developing new methodologies that accurately determine the underlying tissue properties using adaptive beamforming. Accurate estimation of tissue properties will be used to enhance the reconstruction of 3D ultrasound volumes without the need for specialized tracking devices that are bulky and hinder workflow.***Better interpretation will be addressed by using machine learning solutions that automatically identify key anatomical landmarks in 3D ultrasound, segment relevant anatomical boundaries, and fuse 3D ultrasound with pre-operative images to take advantage of complimentary anatomical information. Validation will be established in an iterative design process. An example design application is spinal anesthesia, for guiding epidural needle injection procedures.***All these techniques will be implemented individually or together on commercial point-of-care ultrasound machines starting with the Canadian-made Clarius. These techniques will make several image-guided procedures, including spinal anesthesia, easier and more accurate than current state-of-the-art in these procedures. Hence, this research program will enable widespread deployment of image-guided technology.***The proposed program will train three PhD and three MSc students, as well as five undergraduate students. A detailed training plan that allows individuals to reach their full potential is integrated within the development of research objectives to ensure a high quality, interactive, flourishing, and an equitable research environment for HQP as per UBC Equity and Diversity Strategic Plan and Policy #2.