Epilepsy, sometimes known as seizure disorder, is a brain disorder that affects individuals of all ages. Although it can be seen as a rare disease, there are an estimated 50 million people affected by epilepsy worldwide. In developing countries, such as Guatemala, the diagnosis and treatment of this disease is difficult, due to cultural and socioeconomic factors. Furthermore, the study of epilepsy is limited to a small percentage of neurologists.

Considering these factors, in 2020 the Department of Electronics, Mechatronics and Biomedical Engineering of the Universidad Del Valle de Guatemala (UVG), in collaboration with the Center for Epilepsy and Functional Neurosurgery (HUMANA), started a project that seeks to develop and implement a database and a software tool for the storage, processing and analysis of biomedical signals of patients with epilepsy.

The goal is to apply machine learning and pattern recognition methods to these signals, in order to find important features and detect epileptic episodes automatically.

Great teamwork

software epylepsy

The software tool for the study of Epilepsy

This project is the result of the UVG initiative to collaborate with different organizations to develop projects and research. The first phase was developed as a graduation project of Mechatronics Engineers María Jesús Angulo and María Fernanda Pineda. Angulo focused on the processing and analysis of signals, as well as machine learning algorithms. Pineda focused on the database.

The graduates worked together with HUMANA, the only entity of specialized neurosurgeons for the treatment of Epilepsy and Parkinson’s diseases in Guatemala. The project team also includes Dr. Luis Alberto Rivera, as an advisor. During his studies at the University of Missouri, Dr. Rivera worked on Assistive Technologies projects, applying machine learning to biomedical signals. Ing. Hector Antonio Hurtarte, who teaches courses on databases in the Department of Computer Science and Engineering at UVG, was also an advisor to the project.

A multidisciplinary project

One of the most important challenges for the engineers Angulo and Pineda was to become familiar with topics outside their area: epilepsy, biomedical signals such as electroencephalography (EEG), machine learning algorithms, and databases.

The combined work led to the developed software tool, which integrates a signal processing, analysis and visualization module, and an interface for interaction with a relational database. The database is capable of storing information from tests performed, for reference and posterior analysis.

To validate the tool, the algorithms and database functions were applied to signals taken from public biomedical signals repositories. After the validation, the tool was applied to signals from anonymous HUMANA patients with positive results.

The database is currently being expanded and functionality is being added to the software tool. These will be very useful for HUMANA physicians and anyone interested in the study of epilepsy.