Our autonomic nervous system ensures that we breathe and that the heart keeps on pumping. But sometimes, this cardiovascular system does not function as it should. It may be because of various illnesses or injuries in the body's nerves, blood vessels, various organs or the brain itself. This in turn can lead to a heart rate or blood pressure that shows abnormal values or very irregular variation between heartbeats. Worst case, this can result in a cardiac arrest. If variations and co-variations of heart rate and blood pressure are analyzed, however, dysfunctions of the autonomic nervous system could be detected.
This is where research in Dependable Communication and Computation Systems can contribute. It is about developing a tool for real-time diagnostics by combining existing methods of analysis with artificial intelligence.
– We work with cognitive systems and try to copy the processes as they function in a brain – how representations are formed from a variety of sensors in a brain-like way. The purpose of cognitive systems is to be able to read patterns, patterns that are based on past experience, says Evgeny Osipov.
– Our task is to look at the coding possibilities and to eventually make an autonomous cognitive system. Creating an artificial architecture that can interpret signals in real-time and thus diagnose. The cognitive system could also be used in future applications within e-health.
Better care with automatic system
The solution will be used as the basis for an algorithm that can be able to predict critical events in the cardiovascular system in a fully automatic way and thus contribute to better care for patients with autonomic dysfunctions. Even today, signals are processed fairly accurately, but the technology is limited and the analysis of data from the cardiovascular system is often done by a person. In the future, the analysis may instead be performed by a communication and computing system.
– Of course it’s not about replacing doctors, but to complement their assessments, says Evgeny Osipov.
– We hope that the project’s results will be groundbreaking. And without joint research in computer science and biomedicine it would not be possible. The project leader Docent Urban Wiklund’s expertise on the classical methods of analysis of biomedical signals, is therefore of great importance.
The project is a collaboration between Umeå University and Luleå University of Technology. The project will last for four years and is funded by the Swedish Research Council.