Abstracts

Rationale: The widespread use of commercial EEG reading software is limited by restrictive licenses, local use only, and is constrained by vendor-specific tools. Additionally, current EEG software and atlases for teaching are cost-prohibitive and require local installation of applications. Finally, in resource-limited settings, both clinical care and research in epilepsy are limited by availability of clinical neurophysiology expertise. We aimed to design a web-based platform to overcome these obstacles and allow for greater access to EEG reading, teaching and data exchange.Methods: Our platform is based on readily available open source applications and development tools. It reads and displays EEG data from European Data Format (EDF) files stored on a secure web-server. These files can be uploaded to the web server manually by the user, or automatically by third party EEG recording machines connected and interfacing with the hospital network. A C++ and PHP-based web back-end generates a graphical representation of the EEG data in real-time when requested by the user. The output is sent to the user’s web browser using HTML with embedded SVG vector graphics data and JavaScript. The web front-end running on the user’s computer is implemented using Ajax (asynchronous JavaScript and XML) in order to display the EEG data. Bandwidth limitations do not affect the responsiveness, as EEG pages are generated on the server, and the browser only loads one page at a time.Results: A prototype was implemented on a secure web server. Relying solely on the widely available web standards HTML, JavaScript, and SVG, our solution enables the user to browse and read EEGs using any standard browser without the installation of additional software, and is available on mobile devices such as smart phones and tablets. The user can switch montages, change sensitivities, and add low-pass, high-pass, and notch filters. We have found this prototype capable of accurately displaying data recorded on leading EEG systems. Several users can browse their own EEGs or the same EEG simultaneously. In addition, in a study involving EEG interpretation from University Teaching Hospital in Lusaka, Zambia, 31 pediatric EEGs were successfully uploaded locally, and accessed for interpretation in Boston, MA, USA.Conclusions: Our web-based EEG platform lifts many of the limitations of current EEG software. Uploaded data is available almost instantaneously, and can be browsed and read from any web browser. Future versions will allow for event markers, annotation and reporting, and will include an annotated EEG atlas for reference during development and across ages. Remote reading, teaching, and both national and international EEG data exchange will enable access to expertise across borders, and allow for consultation with peers. Additionally, a globally distributed network of participating EEG readers can contribute to crowd-sourcing of neurophysiological expertise. Source of Funding: World Federation of Neurology