Assessment of user-interaction strategies for neurosurgical data navigation and annotation in virtual reality (2212.08688v1)

Abstract: While virtual-reality (VR) has shown great promise in radiological tasks, effective user-interaction strategies that can improve efficiency and ergonomics are still under-explored and systematic evaluations of VR interaction techniques in the context of complex anatomical models are rare. Therefore, our study aims to identify the most effective interaction techniques for two common neurosurgical planning tasks in VR (point annotation and note-taking) from the state-of-the-arts, and propose a novel technique for efficient sub-volume selection necessary in neuroanatomical navigation. We assessed seven user-interaction methods with multiple input modalities (gaze, head motion, controller, and voice) for point placement and note-taking in the context of annotating brain aneurysms for cerebrovascular surgery. Furthermore, we proposed and evaluated a novel technique, called magnified selection diorama (Maserama) for easy navigation and selection of complex 3D anatomies in VR. Both quantitative and semi-quantitative (i.e., NASA Task Load Index) metrics were employed through user studies to reveal the performance of each interaction scheme in terms of accuracy, efficiency, and usability. Our evaluations demonstrated that controller-based interaction is preferred over eye-tracking-based methods for point placement while voice recording and virtual keyboard typing are better than freehand writing for note-taking. Furthermore, our new Maserama sub-volume selection technique was proven to be highly efficient and easy-to-use. Our study is the first to provide a systematic assessment of existing and new VR interaction schemes for neurosurgical data navigation and annotation. It offers valuable insights and tools to guide the design of future VR systems for radiological and surgical applications.