Validation of High Precision Robot-Assisted Methods for Intracranial Applications: Preliminary Study Abhishek Kaushik 1 , T.A. Dwarakanath 2 , Gaurav Bhutani 3 , Aliasgar Moiyadi 4,5 , Pradip Chaudhari 4 INTRODUCTION Robot-based neurosurgery has the poten- tial to be highly accurate and reliable. The robot is subjected to rigorous tests for validation of safety, accuracy, and practical applicability before animal and human trials. The validation involves stage-wise progressive tests involving (but not limited to) surgical simulation studies 1-4 and experiments on phantoms, 5,6 fruits/ vegetables, 7,8 small animals, 9 and cadavers. 10 We have previously described a novel autonomous neuroregistration technique. In neuroregistration, the surgical robot itself autonomously selects and measures the markers in real patient space and conducts the neuroregistration. Algorithms were developed at our laboratory. Robust marker detection and high precision measurement for anatomic registration were determined using the Taguchi method. Robot-based time-optimal autonomous neuroregistration and neu- ronavigation were used. 5 This paper presents a preliminary study and the results of the robot-based autonomous surgical procedure on a phantom and vegetable specimen. The paper discusses case studies, which simu- lated all aspects of the actual neurosurgical procedure on phantom and vegetable Figure 1. Concentric glass jar phantom with multiple pair of concentric holes. - BACKGROUND: This work attempts to simulate a robot-based autonomous targeted neurosurgical procedure such as biopsy on a vegetable specimen. The objective of the work is to validate the robot-based autonomous neuro- registration and neuronavigation for neurosurgery in terms of stereotactic nav- igation and target accuracy. - CASE DESCRIPTION: A vegetable (carrot) fixed in a tray was used as a model. The tray was affixed with multiple markers. The robot autonomously registers the subject precisely and subsequently accesses the target. The nav- igation trajectory closely follows the path from the entry point to the target point, as specified in the medical image. The replication of procedures reveals that the target accuracies are within 1 mm. The results based on the case studies are presented. Intricate cases in terms of entry hole size, depth, and size of the target are considered for both phantom and vegetable trials. - CONCLUSIONS: The results of the case studies show enhanced and consistent performance characteristics in terms of accuracy, precision, and repeatability with the added advantage of the economy of time. The case studies serve as validation for a high precision robot-assisted neuroregistration and neuronavigation task for neurosurgery and pave the way for further animal and human trials. Key words - Accuracy - Autonomous neuroregistration - Robot-assisted surgery - Validation Abbreviations and Acronyms CT : Computed tomography TRE: Target registration error From the 1 Department of Engineering Sciences, Homi Bhabha National Institute, Mumbai; 2 Division of Remote Handling & Robotics, Bhabha Atomic Research Centre and Homi Bhabha National Institute, Mumbai; 3 Division of Remote Handling & Robotics, Bhabha Atomic Research Centre, Mumbai; 4 Advanced Centre for Treatment, Research and Education in Cancer, Tata Memorial Centre, Mumbai; and 5 Homi Bhabha National Institute, Mumbai, India To whom correspondence should be addressed: Abhishek Kaushik, Ph.D. [E-mail: kaushik.or.abhishek@gmail.com] Supplementary digital content available online. Citation: World Neurosurg. (2020) 137:71-77. https://doi.org/10.1016/j.wneu.2020.01.206 Journal homepage: www.journals.elsevier.com/world- neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2020 Elsevier Inc. All rights reserved. WORLD NEUROSURGERY 137: 71-77, MAY 2020 www.journals.elsevier.com/world-neurosurgery 71 Case Report