Validation of a surgical navigation system for hypertensive intracerebral hemorrhage based on mixed reality using an automatic registration method

Hypertensive intracerebral hemorrhage (HICH) is a kind of intracerebral bleeding disease that affects 2.5 per 10,000 people world wide each year. An effective way of curing this disease is to perform a puncture procedure through the dura with a brain puncture drill and tube. The insertion accuracy determines the quality of surgery. Currently, augmented reality-and mixed reality (MR)-based surgical navigation is a promising new technology for surgical navigation in the clinic, aiming to improve the safety and accuracy of surgery. In this study, we present a novel multimodel MR navigation system for HICH surgery. With multi-information fusion technology, organs and hematoma data extracted from CT images can be fused with real patient, which allows surgeons to perform punctures easily. An automatic registration method with a 3D-printed fiducial marker was performed to significantly decrease the time required for surgery preparation.Phantom experiments and user tests were performed in this study, the results of these phantom experiments demonstrated that the mean registration error was 1.18 mm, the insertion error was 1.74 and the average time consumption was 15.9 min, which indicating that this approach was sufficient for clinical application. All the experimental results indicated that this system shows particular promise for use in training inexperienced surgeons, and the next steps would be to refine the system based on the findings with more experienced surgeons.