Computational Methods Applied in Evaluating the Alterations in Human EEG Activity Caused by ELF Magnetic Field Exposures

Dean Cvetkovic
School of Electrical and Computer Engineering
RMIT University, Melbourne, Australia


An estimated 50 per cent of the population will experience at some time insomnia difficulty falling asleep and remaining asleep resulting in decreased levels of productivity and increasing the risks of accidents and medical abuse. Dr Cvetkovic is currently working to develop a new, drug- free alternative to facilitate relaxation and regulate sleep, that may assist reduce the reliance on hypnotic and sedative medication. In the past, his Ph.D. research work investigated whether various stimuli cause sleep induction. The pilot and main experimental findings suggest that the combination of light, sound and electromagnetic field stimulation could cause changes in brain wave activity characterised with sleep induction.

Dr Cvetkovic's pilot and main study on 40 human subjects, investigated whether human electrical brain-wave or electroencephalographic (EEG) activity could be altered when stimulated by uniform and non-uniform extremely low frequency (ELF) magnetic field. The off-line signal processing and statistical results performed using Matlab and SPSS tools on the recorded EEG data revealed significant changes in particular EEG patters depending on the stimuli frequency. Dr Cvetkovic assumes that the effect in the specific EEG patterns is possibly related to synchronisation, induced rhythmic and synchrony spread theories of neuron firing rate after ELF magnetic field.

These research findings complement present studies and encourage future research in the areas of sleep disorders and electromagnetic bio-effects, using current biomedical engineering approaches. Ultimately through this project Dr Cvetkovic hopes to encourage a collaborative approach to sleep research by multidisciplinary participation of engineering, computer science, neurophysiology and biomedical sciences at a national and international level. The continuation of this research work is already in- place to investigate a possibility of EMF frequency-window effect and a design and development of an automated EEG biofeedback or neurofeedback system that could record and process EEG signals in real- time. This neurofeedback system may include the implementation of various feature extractions, sophisticated neural-fuzzy system, adaptive algorithm and statistical techniques in order to detect various EEG stages associated with human psychological mood states and sleep induction.

Colloquia Series page.