Enhancing Clinical Workflow for Intracranial Seizure Recordings with UI-based Brain Heatmap Visualization
Stephanie Hu1, Raphael Christin2, Jonathan Kleen1
1University of California, San Francisco, 2McGill University
Objective:
To enhance diagnostic interpretation through developing a graphical user interface (UI) for visualizing seizure activity from intracranial recordings on 3-dimensional brain reconstructions in patients with drug-resistant epilepsy.
Background:
Human intracranial recordings (ICEEG) play a critical role in seizure localization, but are often tedious to interpret. This challenge has been met by emerging approaches in data visualization, in which ictal activity is converted and projected as heatmaps on personalized 3-D brain reconstructions. Their use has been clinically validated as an adjunct for detecting seizure-onset zones, but requires computer programming experience. This poses a challenge for end-users (clinicians) and necessitates a UI to increase production speed and broaden accessibility.
Design/Methods:
ICEEG data was collected from patients with drug-resistant epilepsy who underwent intracranial implantation of grid, strip, and depth electrodes between 2016–2025 at a single institution. Imaging-based brain reconstructions were fused with ICEEG of recorded seizures using a UI to streamline layout design for brain heatmap plots. The UI was used to regenerate existing videos and applied to new patients to demonstrate framework adaptability to unseen data, and manual processing time was quantified.
Results:
The UI enabled versatile user configuration of brain surface renderings and signal parameters for seizure-spread visualization. Workflow for rendering seizure heatmap visualizations was reduced from roughly 120 minutes per patient to 20–45 minutes. Separately, video file generation ranged from several minutes to two hours, highly dependent upon seizure duration and the target frame rate. The UI framework also enabled near real-time preview windows of functional activity.
Conclusions:

A user-friendly UI enabled clinicians to efficiently generate video-based visualizations of seizure spread from ICEEG data. This tool is designed to be open-access to enhance availability and integration into clinical workflows where identifying seizure onset zones is critical for patient care. Future work will focus on further clinical validation, including real-time visualization capabilities.


10.1212/WNL.0000000000216759
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.