Automatic Intraoperative Correction of Brain Shift for Accurate Neuronavigation.

Area of Science:

• Neurosurgery
• Medical Imaging
• Image Registration

Background:

• Conventional magnetic resonance imaging (MRI)-guided neurosurgery faces challenges with inaccurate patient-to-image registration and brain shift, compromising neuronavigation reliability.
• A novel system employing intraoperative three-dimensional (3D) ultrasound and MRI-to-ultrasound registration was developed to automatically correct these inaccuracies.

Purpose of the Study:

• To prospectively validate the performance of an automatic neuronavigation correction system during intraoperative neurosurgery.
• To assess the system's accuracy and feasibility in guiding tumor resections.

Main Methods:

• The system was evaluated in 13 neurosurgical tumor resection cases.
• Performance was assessed both during the surgery (intraoperatively) and after the surgery (postoperatively).
• The system utilizes intraoperative 3D ultrasound and MRI-to-ultrasound registration for automatic correction.

Main Results:

• The system provided accurate guidance for tumor resection in 9 out of 13 cases intraoperatively.
• Median alignment error improved from 5.12 mm to 2.72 mm after intraoperative correction using manually placed anatomic landmarks.
• Postoperative analysis identified system limitations that were subsequently modified to achieve sufficient accuracy in all cases.

Conclusions:

• Automatic and accurate correction of spatially unreliable neuronavigation is achievable within surgical constraints.
• The study successfully identified and addressed limitations of the developed system, paving the way for improved neurosurgical navigation.