Richa Rogério, Vágvölgyi Balázs, Balicki Marcin, Hager Gregory D, Taylor Russell H
Johns Hopkins University, USA.
Med Image Comput Comput Assist Interv. 2012;15(Pt 1):397-404. doi: 10.1007/978-3-642-33415-3_49.
Current technical limitations in retinal surgery hinder the ability of surgeons to identify and localize surgical targets, increasing operating times and risks of surgical error. In this paper we present a hybrid tracking and mosaicking method for augmented reality in retinal surgery. The system is a combination of direct and feature-based tracking methods. A novel extension for direct visual tracking using a robust image similarity measure in color images is also proposed. Several experiments conducted on phantom, in vivo rabbit and human images attest the ability of the method to cope with the challenging retinal surgery scenario. Applications of the proposed method for tele-mentoring and intra-operative guidance are demonstrated.
视网膜手术当前的技术局限性阻碍了外科医生识别和定位手术靶点的能力，增加了手术时间和手术失误风险。在本文中，我们提出了一种用于视网膜手术增强现实的混合跟踪与拼接方法。该系统是直接跟踪方法和基于特征的跟踪方法的结合。还提出了一种在彩色图像中使用鲁棒图像相似性度量进行直接视觉跟踪的新颖扩展方法。在模拟模型、活体兔和人体图像上进行的多项实验证明了该方法应对具有挑战性的视网膜手术场景的能力。展示了所提出方法在远程指导和术中引导方面的应用。
