Barbara Carl1, Miriam Bopp2,3, Benjamin Saß2, Mirza Pojskic2, Christopher Nimsky2,3. 1. Department of Neurosurgery, University Marburg, Baldingerstrasse, 35033, Marburg, Germany. carlb@med.uni-marburg.de. 2. Department of Neurosurgery, University Marburg, Baldingerstrasse, 35033, Marburg, Germany. 3. Marburg Center for Mind, Brain and Behavior (MCMBB), Marburg, Germany.
Abstract
BACKGROUND: Microscope-based augmented reality (AR) is commonly used in cranial surgery; however, until recently, this technique was not implemented for spinal surgery. We prospectively investigated, how AR can be applied for intradural spinal tumor surgery. METHODS: For ten patients with intradural spinal tumors (ependymoma, glioma, hemangioblastoma, meningioma, and metastasis), AR was provided by head-up displays (HUDs) of operating microscopes. User-independent automatic AR registration was established by low-dose intraoperative computed tomography. The objects visualized by AR were segmented in preoperative imaging data; non-linear image registration was applied to consider spine flexibility. RESULTS: In all cases, AR supported surgery by visualizing the tumor outline and other relevant surrounding structures. The overall AR registration error was 0.72 ± 0.24 mm (mean ± standard deviation), a close matching of visible tumor outline and AR visualization was observed for all cases. Registration scanning resulted in a low effective dose of 0.22 ± 0.16 mSv for cervical and 1.68 ± 0.61 mSv for thoracic lesions. The mean HUD AR usage in relation to microscope time was 51.6 ± 36.7%. The HUD was switched off and turned on again in a range of 2 to 17 times (5.7 ± 4.4 times). Independent of the status of the HUD, the AR visualization was displayed on monitors throughout surgery. CONCLUSIONS: Microscope-based AR can be reliably applied to intradural spinal tumor surgery. Automatic AR registration ensures high precision and provides an intuitive visualization of the extent of the tumor and surrounding structures. Given this setting, all advanced multi-modality options of cranial AR can also be applied to spinal surgery.
BACKGROUND: Microscope-based augmented reality (AR) is commonly used in cranial surgery; however, until recently, this technique was not implemented for spinal surgery. We prospectively investigated, how AR can be applied for intradural spinal tumor surgery. METHODS: For ten patients with intradural spinal tumors (ependymoma, glioma, hemangioblastoma, meningioma, and metastasis), AR was provided by head-up displays (HUDs) of operating microscopes. User-independent automatic AR registration was established by low-dose intraoperative computed tomography. The objects visualized by AR were segmented in preoperative imaging data; non-linear image registration was applied to consider spine flexibility. RESULTS: In all cases, AR supported surgery by visualizing the tumor outline and other relevant surrounding structures. The overall AR registration error was 0.72 ± 0.24 mm (mean ± standard deviation), a close matching of visible tumor outline and AR visualization was observed for all cases. Registration scanning resulted in a low effective dose of 0.22 ± 0.16 mSv for cervical and 1.68 ± 0.61 mSv for thoracic lesions. The mean HUD AR usage in relation to microscope time was 51.6 ± 36.7%. The HUD was switched off and turned on again in a range of 2 to 17 times (5.7 ± 4.4 times). Independent of the status of the HUD, the AR visualization was displayed on monitors throughout surgery. CONCLUSIONS: Microscope-based AR can be reliably applied to intradural spinal tumor surgery. Automatic AR registration ensures high precision and provides an intuitive visualization of the extent of the tumor and surrounding structures. Given this setting, all advanced multi-modality options of cranial AR can also be applied to spinal surgery.
Authors: Fabian Sommer; Ibrahim Hussain; Sertac Kirnaz; Jacob L Goldberg; Rodrigo Navarro-Ramirez; Lynn B McGrath; Franziska A Schmidt; Branden Medary; Pravesh Shankar Gadjradj; Roger Härtl Journal: Neurospine Date: 2022-09-30
Authors: Alicia Pose-Díez-de-la-Lastra; Rafael Moreta-Martinez; Mónica García-Sevilla; David García-Mato; José Antonio Calvo-Haro; Lydia Mediavilla-Santos; Rubén Pérez-Mañanes; Felix von Haxthausen; Javier Pascau Journal: Sensors (Basel) Date: 2022-06-29 Impact factor: 3.847