Multimodal navigated skull base tumor resection using image-based vascular and cranial nerve segmentation: A prospective pilot study

Abstract

Background:Skull base tumors frequently encase or invade adjacent normal neurovascular structures. For this reason, optimal tumor resection with incomplete knowledge of patient anatomy remains a challenge.Methods:To determine the accuracy and utility of image-based preoperative segmentation in skull base tumor resections, we performed a prospective study. Ten patients with skull base tumors underwent preoperative 3T magnetic resonance imaging, which included thin section three-dimensional (3D) space T2, 3D time of flight, and magnetization-prepared rapid acquisition gradient echo sequences. Imaging sequences were loaded in the neuronavigation system for segmentation and preoperative planning. Five different neurovascular landmarks were identified in each case and measured for accuracy using the neuronavigation system. Each segmented neurovascular element was validated by manual placement of the navigation probe, and errors of localization were measured.Results:Strong correspondence between image-based segmentation and microscopic view was found at the surface of the tumor and tumor-normal brain interfaces in all cases. The accuracy of the measurements was 0.45 ± 0.21 mm (mean ± standard deviation). This information reassured the surgeon and prevented vascular injury intraoperatively. Preoperative segmentation of the related cranial nerves was possible in 80% of cases and helped the surgeon localize involved cranial nerves in all cases.Conclusion:Image-based preoperative vascular and neural element segmentation with 3D reconstruction is highly informative preoperatively and could increase the vigilance of neurosurgeons for preventing neurovascular injury during skull base surgeries. Additionally, the accuracy found in this study is superior to previously reported measurements. This novel preliminary study is encouraging for future validation with larger numbers of patients.

Multimodal Navigation in Endoscopic Transsphenoidal Resection of Pituitary Tumors Using Image-Based Vascular and Cranial Nerve Segmentation: A Prospective Validation Study

“…Our group achieved a measurement accuracy of 1.20 + 0.21 mm (mean + SD), which is comparable to previously reported accuracies ranging from 0.71 to 5 mm 14,17,24,25 . Additionally, we utilized preoperative 3TMRI for image acquisition, which is a far less invasive procedure than DSA.…”

“…Using these images, we can see very small cranial nerves passing in the cisterns or in close vicinity to a tumor. However, in the current study and our previous reports 24, 25 , we showed how we can upload these images into the navigation system, merge them together, segment cranial nerves and the tumor or adjacent blood vessels and use them in real time during the operation to prevent damage to neurovascular elements. Therefore, the major advantage of our method is actual clinical application of these images in the operating room and inside the cranium.…”

“…Image-based vascular and cranial nerve 3D segmentation was found to have an accuracy of 0.45 + 0.21 mm (mean + standard deviation) for microvascular decompression, with excellent correspondence found between image-based segmentation and endoscopic and microscopic images and videos 24 . Additionally, 3D reconstruction was found to have an accuracy of 0.45 + 0.21 mm (mean + standard deviation) for skull base tumor resection, and was highly informative preoperatively and could increase the vigilance of neurosurgeons for preventing neurovascular injury during skull base operations 25 .…”

Augmented reality in neurosurgical navigation: A survey

“…Dolati et al 34 AR MRI P/I Determined the accuracy and utility of image-based preoperative segmentation in skull base tumour resections, we performed a prospective study.…”

Image guidance in skull base tumor resection: A synergistic approach using intraoperative navigated angiosonography for real-time vessel visualization

“…We read with great interest the article by Parviz Dolati et al .,[ 1 ] entitled “Multimodal navigated skull base tumor resection using image-based vascular and cranial nerve segmentation: A prospective pilot study.”…”