Bioelectrical signals interpretation
State-financed long-term Project Reg nr. 0142084As02
D.Sc., Ph.D. Hiie Hinrikus
Report 2005
1. Bioelectrical signals interpretation: main results
The main attention was aimed to EEG analyses to distinguish effects and possible interaction mechanisms of weak (microwave radiation) and strong (anesthetic drugs) stressors on human brain.
Studies on analysis of pulse wave parameters were aimed to noninvasive indirect estimation of elastic properties of blood vessels. Analysis of pulse wave shape is promising as a measure of arterial elasticity and stiffness.
Digital analysis of the spectrophotometrical signal and on-line processing of its components during hemodialysis was aimed to develop a method for monitoring of dialysis treatment and improve dialysis adequacy and quality.
1.1. Microwave effects on EEG
1.1.1. Modulated microwave radiation at low non-thermal level of field power density can affect human central nervous system in a sensible way. During this year, the study continued on discrimination of changes, produced by low-level microwave exposure in intensity and time variability of the human EEG at rest. The power spectral density (PSD) method and nonlinear scaling analysis of the length distribution of low variability periods (LDLVP) were selected for analysis of the EEG signal. During the experiment, 19 healthy volunteers were exposed to a microwave (450 MHz) of 217 Hz frequency (as most characteristic for mobile phones) on-off modulation. The field power density at the scalp was 0.16 mW/cm2. The experimental protocol consisted of ten cycles of repetitive microwave exposure. Signals from frontal, temporal, parietal and occipital EEG channels on EEG theta, alpha and beta rhythms were analysed.
Exposure to microwave causes average increase of EEG activity. LDLVP analysis discriminated significant effect in time variability for 2 subjects. PSD method detected significant changes in intensity for 4 subjects. The effect of low-level microwave exposure is stronger on EEG beta rhythm in temporal and parietal regions of the human brain. The analysis by the LDLVP and PSD methods detected the effect of exposure for about 11% and 21 % of subjects respectively. For instance, the rate of multiple chemical sensitivity (MCS) occurrence is estimated to be between 2 and 10 % in the general population. Taking this into consideration, low-level microwave exposure can influence even higher part of population than multiple chemically unrelated compounds. (M.Bachmann, J.Kalda, M.Säkki, J.Lass, V.Tuulik, H.Hinrikus).
1.1.2. A possible origin of interaction mechanism of microwave radiation with nervous system – quasi-thermal field effect – was investigated. The high frequency microwave field cannot cause any regular changes in the ions movement due to their small absorption cross-section as well as inertial properties and viscosity of the liquid medium. However, the microwave field can cause fluctuations and vibration of the charged particles and membranes in tissues. This phenomenon is similar to the effect caused by Brown motion initiated by temperature and results in the same effects without rise in temperature. The electric field of 1 V/cm can introduce disturbance of the thermal equilibrium inside a cell of 10 ?m radius, which is equivalent to disturbance produced by temperature rise of 1 K. The hypothesis, that microwave heating should cause an effect independent of the microwave modulation frequency, while non-thermal field effect depends on modulation frequency, was examined experimentally. The 450 MHz microwave radiation, modulated at 7, 14 and 21 Hz frequencies, power density at the skin 0.16 mW/cm2, was applied. The experimental protocol consisted of two series of five cycles of the repetitive microwave exposure at fixed modulation frequencies. Relative changes in EEG theta, alpha and beta rhythms of the group of 13 healthy volunteers were analysed. Analysis of the experimental data shows that: 1) statistically significant changes in EEG rhythms depend on modulation frequency of the microwave field; 2) microwave stimulation causes an increase of the EEG energy level; 3) the effect is most intense at beta1 rhythm and higher modulation frequencies. These findings confirm the quasi-thermal origin of the effect, different from average heating. (H.Hinrikus, M.Bachmann, J.Lass, R.Tomson).
1.1. EEG in anesthesia
Our interest has recently been to compare the behavior of the various entropy/complexity measures at different levels of sedation in the ICU. We have found that the various ways of quantifying signal entropy/complexity depend on different signal properties causing sometimes their contradictory behavior. Main problem we met in comparing the entropy/complexity measures of the EEG signal was the dependence of spectral entropy on the length of the signal window. During this year, our aim was to test if this dependence was due to added information when incorporating more data or if it was due to the algorithm.
The EEG data recorded from 12 ICU patients were analyzed using four different schemes of power spectrum estimation for obtaining spectral entropy. Two of the schemes comprise the Welch periodogram averaging method, one scheme is based on the estimation of the autocorrelation function and one on the autoregressive modelling. The results show that spectral entropy values depend highly on the smoothness of the power spectrum estimate. Spectral entropy correlates significantly with data length only if FFT is used for power spectrum estimation and the FFT size varies together with the data length.
The conclusions can be drawn from the analysis presented:
- the ability of the estimate of EEG spectral entropy to differentiate between various levels of sedation does not depend on the method used for power spectrum estimation in general. However, spectral entropy values achieved using different methods for power spectrum estimation are not comparable with each other;
- in the case of periodogram averaging, higher spectral entropy estimates for longer data windows are not due to the additional information contained in the data but rather comes from high variance of the power spectrum estimate typical to this method;
- in general, smoother power spectrum estimates (using AR-model coefficients, for example) give higher values of spectral entropy. ( R.Ferenets, A.Anier).
1.3. Noninvasive monitoring of arterial elasticity
Impaired vascular compliance and a concurrent rise in vascular rigidity are the central pathogenetic processes and the first step leading to fatal cardiovascular events in many cases of hypertension and hyperlipidaemia patients. Among the noninvasive methods of evaluating arteries, pulse wave shape analysis can be used as an index of arterial elasticity and stiffness. The goal of the study was twofold: 1) to evaluate whether pulse wave shape is modified by cold pressure test (CPT), and 2) compare the changes in the radial arterial pulse shape during CPT in normotensive patients and in patients with borderline hypertension.
During this study the changes in arterial pulse wave shape for 21 patients (aged 22 to 64 years) were investigated. Measurements of pulse wave shape were performed in normal conditions and with applying CPT. The results showed that pulse wave shape was significantly changed with increasing of blood pressure. The CPT caused statistically significant changes in pulse wave shape in the case of patients with normal blood pressure and did not induced significant change for the patients with hypertension. The preliminary results show that it is possible to use pulse wave shape analysis for noninvasive indirect estimation of elastic properties of blood vessels. Cooperation with North Estonia Regional Hospital. (I.Hlimonenko, K.Meigas, R.Vahisalu)
1.4. Analysis of the spectrophotometrical signal in hemodialysis
The aim of the research is to develop a new technique to estimate dialysis adequacy and quality. The method is based on UV-absorption phenomena enabling on-line monitoring of solutes in the spent dialysate. The technique offers a possibility to follow haemodialysis session continuously, monitor deviations in dialysis efficiency, and estimate the quality and adequacy of the dialysis.
During this year, a preliminary validation of the method was performed. The study was aimed to compare the total removed uric acid obtained from the on-line UV-absorbance measurements in the spent dialysate (TRua) and the total dialysate collection (TDC) as reference method. Six uremic patients, two females and four males, on chronic thrice-weekly hemodialysis were included in the study. All spent dialysate during dialysis was collected in a tank that gave the TDC value. A double-beam spectrophotometer was used for the determination of UV-absorbance. Regression line from the first in week sessions was assessed to transform UV-absorbance into uric acid concentration. This relationship was used for the subsequent treatments’ TRua calculations for each patient (method UV1). Also, TRua based on the transformation using the regression line from the total material was calculated (method UV2). TRua from the three methods was finally compared. TRua obtained using TDC and two different transformations for UV-absorbance (mean ± SD) in ?mol were: 5277 ± 897 from TDC (N = 23), 5150 ± 882 from UV1 (N = 23), and 5422 ± 1112 from UV2 (N = 23) transformation. None of the mean TRua values were significantly different (P > 0.05).
The results show the possibility to estimate total removed uric acid by using UV-absorbance. Still a larger material is needed to draw more general conclusions. Cooperation with Institute of Biomedical Engineering, Linköping University, and North Estonia Regional Hospital. (I. Fridolin, M.Luman)
2. Publications
Papers refereed
1. Bachmann M., Kalda J., Lass J., Tuulik V., Sakki M., Hinrikus H. Non-linear analysis of the electroencephalogram for detecting effects of low-level electromagnetic fields. Med. Biol. Eng. Comput. 43, 142-149, 2005.
2. M.Bachmann, M.Säkki, J.Kalda, J.Lass, V.Tuulik, H.Hinrikus. Effect of 450 MHz microwave modulated with 217 Hz on human EEG in rest. The Environmentalist, 25, 165-171, 2005.
3. H.Hinrikus, R.Tomson, J.Lass, D.Karai, J.Kalda, V.Tuulik. Low-level microwave radiation effect on nerve pulse conduction velocity. The Environmentalist, 25, 157-163, 2005.
4. Rodina A., Lass J., Riipulk J., Bachmann T., Hinrikus H. Study of effects of low level microwave field by method of face masking. Bioelectromagnetics, 26, 571-577, 2005.
5. H.Hinrikus, M.Bachmann, R.Tomson, J.Lass. Non-thermal effect of microwave radiation on human brain. The Environmentalist 25, 187-194, 2005.
6. Uhlin, F., Fridolin I., Lindberg LG, Magnusson M. Estimating total urea removal and protein catabolic rate by monitoring UV absorbance in spent dialysate. Nephrol Dial Transplant. Epub ahead of print, PubMed.,2005.
7. R. Tomson, H.Hinrikus, M. Bachmann, J.Lass and V.Tuulik. Sensitivity of human EEG to modulation frequency of microwave radiation. 3rd European Medical & Biological Engineering Conference EMBEC´05, Nov. 20-25, 2005, Prague, Czech Republic, IFMBE Proceedings, 11, CD-ISSN: 1727-1983, 5 pages, 2005.
8. R.Ferenets, T. Lipping, P. Puumala, V. Jäntti, E. Sonkajärvi, E. Heikkinen, E. Karvonen, and K. Suominen. Comparing Entropy/Complexity of EEG during Propofol and Sevoflurane Anesthesia at Burst–Suppression Level. Proceedings of the 5th International Workshop on Biosignal Interpretation, September 6-8, Tokyo, Japan, 187-190, 2005.
9. T. Lipping, R. Ferenets, A. Anier, S. Melto, S. Hovilehto. Power spectrum estimation in the calculations of spectral entropy to assess depth of sedation, 3rd European Medical & Biological Engineering Conference EMBEC´05, Nov. 20-25, 2005, Prague, Czech Republic, IFMBE Proceedings, 11, CD-ISSN: 1727-1983, 4 pages, 2005.
10. K. Meigas, J. Lass, D. Karai, R. Kattai, J. Kaik. Device for Continuous Blood Pressure Measurements. 13th Nordic Baltic Conference on Biomedical Engineering and Medical Physics, June 13-17, 2005, Umeå, Sweden., IFMBE Proceedings 9, pp.164-165, 2005.
11. K. Meigas. J. Lass, D. Karai, R. Kattai, J. Kaik. Method and Device for Beat-to-beat Blood Pressure Measurements. 3rd European Medical & Biological Engineering Conference EMBEC´05, Nov. 20-25, 2005, Prague, Czech Republic, IFMBE Proceedings, 11, CD-ISSN: 1727-1983, 6 pages, 2005.
12. Irina Hlimonenko, Kalju Meigas, Rein Vahisalu. Change of Arterial Pulse Wave in Patients with Hyperlipidaemia. 13th Nordic Baltic Conference on Biomedical Engineering and Medical Physics, June 13-17, 2005, Umeå, Sweden., IFMBE Proceedings 9, pp.160-161, 2005.
13. Irina Hlimonenko, Kalju Meigas, Rein Vahisalu. Arterial Pulse Wave Analysis During Cold Pressor Test in Patients with Borderline Hypertension. 3rd European Medical & Biological Engineering Conference EMBEC´05, Nov. 20-25, 2005, Prague, Czech Republic, IFMBE Proceedings 11, CD-ISSN: 1727-1983, 5 pages, 2005.
14. Uhlin, F., I. Fridolin, et al. Response to Clearance Reduction During Haemodialysis by Online UV-Absorbance Monitoring. 13th Nordic Baltic Conference on Biomedical Engineering and Medical Physics, June 13-17, 2005, Umeå, Sweden., IFMBE Proceedings 9, pp.174-175, 2005.
15. Uhlin, F., L. G. Lindberg, I. Fridolin. Total Removed Uric Acid During Dialysis Estimated by On-line Ultra Violet Absorbance in the Spent Dialysate. 3rd European Medical & Biological Engineering Conference EMBEC´05, Nov. 20-25, 2005, Prague, Czech Republic, IFMBE Proceedings 11, CD-ISSN: 1727-1983, 6 pages, 2005.
Abstracts
1. Uhlin, F., I. Fridolin, et al. On-Line Monitoring of the Spent Dialysate During Haemodialysis Using UV-Absorbance. 34th International Conference of EDTNA/ERCA, Wien, Austria.
Submitted
1. R. Ferenets, T. Lipping, A. Anier, V. Jäntti, S.Melto, S. Hovilehto. Comparison of Entropy and Complexity Measures for the Assessment of Depth of Sedation, accepted to IEEE Transactions on BME.
Conferences, workshops
1. 13th Nordic Baltic Conference on Biomedical Engineering and Medical Physics, June 13-17, 2005, Umeå, Sweden: K.Meigas, I.Fridolin, I.Hlimonenko.
2. 5th International Workshop on Biosignal Interpretation, September 6-8, Tokyo, Japan: R.Ferenets.
3. 3rd European Medical & Biological Engineering Conference EMBEC´05, Nov. 20-25, 2005, Prague, Czech Republic: H.Hinrikus, K.Meigas, I.Fridolin, I.Hlimonenko, V.Tuulik.
4. Supportive grants
1. Ruth Tomson, Estonian Ministry of Education and Research, Kristjan Jaak grant for MSc study at MälardalenUniversity, Sweden, Sept 15 - Dec. 31.
2. Rain Ferenets. Estonian Ministry of Education and Research, Kristjan Jaak grant for doctoral study and research at Tampere University of Technology, April 01 – Dec. 31.
3. Anastassia Rodina, State Scholarships Foundation, Greece, grant for PhD study at University of Patras, Oct. 2005-2006.
5. International Projects and Programmes
1. COST 281 Potential Health Implications from Mobile Communication Systems
2. COST B27 Electric neuronal oscillations and cognition (ENOC).
3. NATO grant Optical methods for diagnosis and monitoring of clinical parameters, I. Fridolin
National:
Estonian Programme on Biomedical Engineering
6. Visiting scolars and researhers
1. Prof. Jaakko Malmivuo, Ragnar Granit Institute, Tampere University of Technology, Finland, Intensive course “Bioelectromagnetism” April 18-22.
2. Prof. Tarmo Lipping, Sen. Res. Ville Jäntti, Tampere University of Technology, workshop “EEG in intensive care unit”, March 03.
7. Visiting students
1. Giovanni Nolli, Italy, MSc student, Sept-Dec.
2. Fernando Perez, Spain, PhD student, Sept-Dec.
8. Theses
Promoted:
MSc theses
Irina Hlimonenko Pulse Wave Analysis in Patients with Hypertension and Hyperlipidaemia, supervisor K. Meigas
BSc theses
Raul Kala, Anatoli Vaiser, Jekaterina Rubljova, Ketlin Ilau, Jana Jerotskaja, Irina Krotova, Marina Tukardina, Anna Suhhova, Ain Suik, Alla Janson
In progress
PhD
Maie Bachmann (supervisor H.Hinrikus)
Rain Ferenets (supervisor T.Lipping)
Andres Anier (supervisor T.Lipping)
Irina Hlimonenko (supervisor K.Meigas)
Ellen Suurküla (supervisor I.Fridolin)
MSc
Ruth Tomson (supervisor H.Hinrikus)
Anna Suhhova (supervisor M.Bachmann)
Jekaterina Rubljova (supervisor M.Bachmann)
Jana Jerotskaja (supervisor I.Fridolin)
Lemme Berkis, (supervisor K.Meigas)
Aleksei terbakov (supervisor I.Fridolin)
Gert Vaik (supervisor I.Fridolin)
Kai Lauri (supervisor I.Fridolin)
Aleksei Stserbakov (supervisor I.Fridolin)
Aile Kaasik (supervisor J.Lass)
Kristina Temitski (supervisor J.Lass)
Ain Suik (supervisor K. Meigas)
Irina Krotova (supervisor V.Tuulik)
9. Awards
Rain Ferenets, Young scientist award at IEEE EMBS 5th International Workshop on Biosignal Interpretation, September 6-8, Tokyo, Japan
|
