It is the details that make all the difference. Epileptic disorders can vary greatly in terms of type, cause and severity. This makes having accurate knowledge of the person’s individual medical condition very important when choosing a custom-fit therapy. To date, physicians have to rely on accounts by the person concerned and those close to them when estimating how often seizures occur. However, this information is often incomplete, as seizures are not always consciously perceived. The partners in the »MOND« project are working on a solution: A mobile neurosensing system suitable for everyday use that detects epileptic seizures automatically and documents them for the purpose of medical anamnesis and optimisation of patient safety. The project, which started in April 2020, is coordinated by the Division Hearing, Speech and Audio Technology of the Fraunhofer Institute for Digital Media Technology IDMT in Oldenburg.
The catchy title »MOND« stands for the project’s clear objective: The conceptual development of a »Mobile, Smart Neurosensoring System for the Detection and Documentation of Epileptic Seizures in Daily Life«. A fully documented clinical picture of epilepsy patients would be a major success because it would facilitate highly accurate diagnosis and treatment. »The affected persons themselves, however, often gauge the frequency and intensity of their seizures incorrectly. There can be various reasons for this. Symptoms cannot always be clearly attributed or remain unnoticed, for example, when the person is asleep. We estimate that only 50 % of seizures at most are consciously perceived,« explains Professor Rainer Surges, Director of the Department of Epileptology at the University Hospital of Bonn.
So far, an automatic, mobile documentation of epileptic seizures outside the hospital is not reliably possible. Such an approach would, however, close existing gaps in patients’ medical records and when assessing a therapy’s success. With the analysis of an electroencephalograph (EEG), i.e. of the brain’s electrical activity, recorded in mobile mode, the »MOND« project consortium is treading new and promising ground. This is, however, not without major challenges. The EEG signal itself and its interpretation are complex. Any movement or the slightest muscle contraction (eye movement, talking, walking) causes massive signal interference or produces false measurement results. In the framework of the project, the consortium wants to tackle these false results, known as »artefacts«, using artificial intelligence methods.