Abstracts

Rationale: Source reconstruction algorithms have recently been applied to stereo-electroencephalography (sEEG) signals where the electrodes are implanted into the brain and in close proximity to active neural sources. These approaches assumed that the neural sources generating the sEEG signals can be localized using recordings from a single snapshot in time. However, interictal spikes and seizures are dynamic and propagate from one brain location to another over time. Additionally, there is information at one time point that can inform subsequent localizations. Therefore, we developed a propagating source reconstruction algorithm that uses the recorded signal dynamics to reconstruct moving sources.

Methods: We developed a propagating source reconstruction algorithm based on the static iterative reweighting edge sparse source reconstruction algorithm (IRES). IRES is a convex optimization algorithm that aims to minimize the strength and size of the source while ensuring that the reconstructed recordings do not differ greatly from the true recordings. We used IRES to reconstruct independently sources at every time point of our sEEG recordings, which resulted in reconstructions that were in the neighborhood of the true sources. We then clustered sources that were spatially connected and temporally overlapping to determine the trajectory of the source using each source’s spatiotemporal center of mass. IRES reconstructions had many spurious and spatially discrete sources resulting in many trajectories. We chose the trajectory that was active for the longest time and omitted trajectories that came within 4 cm of this trajectory. We smoothed the trajectory using a sliding window average with window size of 100 time points and generated search regions with a 4 cm radius centered at the trajectory. Since IRES is a convex optimization algorithm, initializing the algorithm well can greatly improve performance. Therefore, we used the trajectory defined regions to initialize a second run of the IRES algorithm to determine the final reconstructions of the moving sources. We created synthetic sources in three patient-specific head models with a left parieto-occipital implantation, a left temporal lobe implantation, and a bilateral exploratory implantation, where electrodes were placed throughout the brain. We created moving synthetic sources in the region of interest for each patient and in the frontal lobe for the bilaterally implanted patient.

Results: We were able to reconstruct the moving synthetic source in each patient using our propagating source reconstruction algorithm. The propagating source reconstruction algorithm outperformed IRES on metrics of jaccard index (0.28 vs. 0.13), percent overlap with true source (28% vs. 60%), and localization error (13.1 mm vs. 7.6 mm).

Conclusions: Propagating source reconstruction improves the reconstruction of neural sources relative to static reconstructions, and our algorithm is generalizable across patients with varied implantation schemes.

Funding: Duke MEDx and NIH F31 NS124094