Magnetoencephalography Atlas Viewer for Dipole Localization and Viewing

Area of Science:

• Neuroimaging
• Medical Physics
• Computational Neuroscience

Background:

• Magnetoencephalography (MEG) is crucial for epilepsy and tumor mapping, but requires expert anatomical localization of dipoles.
• Current clinical reporting for MEG is time-consuming due to the need for multidisciplinary expert input on dipole localization.
• Accurate anatomical localization of MEG dipoles is essential for precise diagnosis and treatment planning.

Purpose of the Study:

• Introduce the Magnetoencephalography Atlas Viewer (MAV), a novel tool to streamline MEG dipole anatomical analysis.
• To develop a user-friendly graphical user interface (GUI) for interactive display of MEG dipoles and their anatomical locations.
• To validate the accuracy of the MAV tool in localizing MEG dipoles compared to expert consensus.

Main Methods:

• The MAV tool normalizes patient Magnetic Resonance Imaging (MRI) to standard MNI space and reverse-normalizes MNI atlases to the native MRI.
• It identifies MEG dipole files and matches their coordinates to spatial locations within atlas files.
• The tool was evaluated on 273 dipoles from clinical epilepsy subjects, with ground truth established by three neuroradiologists.

Main Results:

• The MAV achieved a concordance rate of 79% to 84% with expert ground truth, varying by normalization method.
• Higher concordance (80%-90%) was observed in subjects with minimal or no structural MRI abnormalities.
• The MAV demonstrated a user-friendly GUI for displaying dipoles, coordinates, anatomical labels, and MRI overlays.

Conclusions:

• The Magnetoencephalography Atlas Viewer (MAV) offers a straightforward method for MEG dipole anatomical localization.
• MAV facilitates clinical reporting by enabling nonspecialists to prepopulate reports, significantly reducing reporting time.
• This tool enhances the efficiency and accessibility of MEG data interpretation in clinical settings.