Repeatability of knee impulsive loading measurements with skin-mounted accelerometers and lower limb surface electromyographic recordings during gait in knee osteoarthritic and asymptomatic individuals.

OBJECTIVES: To determine the repeatability of knee joint impulsive loading measurements with skin-mounted accelerometers (SMAs) and lower limb surface electromyography (EMG) recordings during gait. METHODS: Triaxial SMA and EMG from 4 muscles during level and stair walking in nine healthy and nine knee osteoarthritis (OA) subjects were used. The initial peak acceleration (IPA), root mean square (RMS), maximal acceleration transient rate (ATRmax) and mean EMG activity (EMGact) were calculated. The coefficient of variation (CV) and the intraclass correlation coefficient (ICC) were calculated to measure repeatability. RESULTS: The CV and ICC of RMS accelerations ranged from 4.9% to 10.9% and from 0.69 to 0.96 in both study groups during level walking. The CV and ICC of IPA and ATRmax varied from 7.7% to 14.2% and from 0.85 to 0.99 during level and stairs up walking in healthy subjects. The CV and ICC of EMGact ranged from 8.3% to 31.7% and from 0.16 to 0.97 in both study groups. CONCLUSIONS: RMS accelerations exhibited good repeatability during walking in healthy and knee OA subjects. The repeatability of EMG measurements was acceptable in healthy subjects depending on the measured muscles.

Principal Component Regression on Motor Evoked Potential in Single-Pulse Transcranial Magnetic Stimulation.

Motor evoked potentials (MEPs) induced by transcranial magnetic stimulation (TMS) are commonly characterized only by their onset (latency) and size (amplitude) whereas other potentially important information in the MEPs is discarded. Hence, our aim was to examine the morphological information of MEPs using principal component regression (PCR) providing additional perception of MEPs. MEPs were recorded from the first dorsal interosseous muscle following navigated TMS focused at the primary motor cortex. The PCR holding of at least 96% of total variance of the MEP dataset was performed to parameterize MEPs into principal components (PCs), which were used with non-linear least square estimation to reconstruct original MEPs. The comparison between the original and reconstructed MEPs showed that PCs, which accounted for 96% of total variance, were able to characterize the MEP morphology, i.e., the PCR summarizes the repeated information in the MEP dataset into the PC set. In addition, PCR benefited the automated quantification of MEP features as it removed the random noise caused by the environmental interference and the inconsistency of neuronal pathways. Furthermore, we could determine the minimum number of trials required to reliably represent the whole dataset by estimating the partial information of those trials accounted for. Our results showed that this partial information exponentially increased with respect to the number of trials, and saturated within 20 MEPs holding approximately 90% of total variance of the dataset.

A new computational approach for cortical imaging.

Estimation of current or potential distribution on the cortex is used to obtain information about neural sources from the scalp recorded electroencephalogram. If the active sources in the brain are superficial, the estimated field distribution on the cortex also yields information about the active source configuration. In these cases, these methods can be used as source localization methods. In this study, we concentrate on finite-element-based cortex potential estimation. Usually these methods require surface interpolation of the recorded voltages at the electrodes onto the entire scalp surface. We propose a new computational approach which does not require the use of surface interpolation but does it implicitly and uses only the recorded data at the electrodes. We refer to this method as the systematic approach (SA). We compare the SA with the surface interpolation approach (IA) and show that the SA is able to produce somewhat better accuracy than the IA. However, the main asset is that the sensitivity of the cortical potential maps to the regularization parameter is significantly lower than with the IA.

Tikhonov regularization and prior information in electrical impedance tomography.

The solution of impedance distribution in electrical impedance tomography is a nonlinear inverse problem that requires the use of a regularization method. The generalized Tikhonov regularization methods have been popular in the solution of many inverse problems. The regularization matrices that are usually used with the Tikhonov method are more or less ad hoc and the implicit prior assumptions are, thus, in many cases inappropriate. In this paper, we propose an approach to the construction of the regularization matrix that conforms to the prior assumptions on the impedance distribution. The approach is based on the construction of an approximating subspace for the expected impedance distributions. It is shown by simulations that the reconstructions obtained with the proposed method are better than with two other schemes of the same type when the prior is compatible with the true object. On the other hand, when the prior is incompatible with the true object, the method will still give reasonable estimates.

Estimation of event-related synchronization changes by a new TVAR method.

The modeling of nonstationary electroencephalogram (EEG) with time-varying autoregressive (TVAR) models is discussed. The classical least squares TVAR approach is modified so that prior assumptions about the signal can be taken into account in an optimal way. The method is then applied to the estimation of event-related synchronization changes in the EEG. The results show that the new approach enables effective estimation of the parameter evolution of the time-varying EEG with better time resolution compared to previous methods. The new method also allows single-trial analysis of the event-related synchronization.

A Kalman filter approach to track fast impedance changes in electrical impedance tomography.

In electrical impedance tomography (EIT), an estimate for the cross-sectional impedance distribution is obtained from the body by using current and voltage measurements made from the boundary. All well-known reconstruction algorithms use a full set of independent current patterns for each reconstruction. In some applications, the impedance changes may be so fast that information on the time evolution of the impedance distribution is either lost or severely blurred. In this paper, we propose an algorithm for EIT reconstruction that is able to track fast changes in the impedance distribution. The method is based on the formulation of EIT as a state-estimation problem and the recursive estimation of the state with the aid of the Kalman filter. The performance of the proposed method is evaluated with a simulation of human thorax in a situation in which the impedances of the ventricles change rapidly. We show that with optimal current patterns and proper parameterization, the proposed approach yields significant enhancement of the temporal resolution over the conventional reconstruction strategy.

Subspace regularization method for the single-trial estimation of evoked potentials.

A method for the single-trial estimation of the evoked potentials is proposed. The method is based on the so-called subspace regularization approach in which the second-order statistics of the set of the measurements is used to form a prior information model for the evoked potentials. The method is closely related to the Bayesian estimation. The performance of the proposed method is evaluated using realistic simulations. As a specific application the method is applied to the estimation of the target responses in the P300 test.

Analysis of galvanic skin responses with principal components and clustering techniques.

An advanced method for analyzing the patterning of successive galvanic skin responses (GSR) is presented. The proposed method is based on principal component analysis in which the vector containing the measured signal is presented as a weighted sum of orthogonal basis vectors. The method is tested using measurements from 20 healthy controls and 13 psychotic patients. For each subject, 11 surprising auditory stimuli were delivered to right ear at irregular intervals and evoked GSRs were recorded from the hand. For most of the healthy controls, there was a clear pattern in successive GSRs, whereas within psychotic patients the lack of time-locking of GSRs seemed to be characteristical. These between group differences can be revealed by the proposed method. With application to clustering a significant discrimination, with overall correct ratings of 82%, of healthy controls and psychotic patients is achieved. A significant fact is that all patients were ranked correctly giving the proposed method a sensitivity of 100%.

Assessment of errors in static electrical impedance tomography with adjacent and trigonometric current patterns.

In electrical impedance tomography (EIT), difference imaging is often preferred over static imaging. This is because of the many unknowns in the forward modelling which make it difficult to obtain reliable absolute resistivity estimates. However, static imaging and absolute resistivity values are needed in some potential applications of EIT. In this paper we demonstrate by simulation the effects of different error components that are included in the reconstruction of static EIT images. All simulations are carried out in two dimensions with the so-called complete electrode model. Errors that are considered are the modelling error in the boundary shape of an object, errors in the electrode sizes and localizations and errors in the contact impedances under the electrodes. Results using both adjacent and trigonometric current patterns are given.

Estimation of the dynamics of event-related desynchronisation changes in electroencephalograms.

A method for the estimation of medium rate transitions of non-stationary electroencephalograms (EEG) is proposed. The method is applicable to such EEG dynamics that are between (a) fast transitions for which segmentation procedures are used and (b) slow transitions for which adaptive filters work properly. The estimation of the transition dynamics is based on a novel time-varying autoregressive model. This model belongs to the class of deterministic regression time-varying autoregressive models and its parametrisation allows only simultaneous transitions in all coefficient evolutions. Data from 22 patients was analysed. The performance of the method is first evaluated with realistic simulations of known transition dynamics and it is shown to be able to track medium-rate transitions. The method is then applied to the estimation of the dynamics of event related desynchronisation. It is shown that the proposed method is able to estimate the transitions which are less apparent, such as from a multi-infarct patient.

Effects of electrode properties on EEG measurements and a related inverse problem.

A trend in EEG measurements is to increase the number of measurement electrodes in order to improve the spatial resolution of the recorded voltage distribution at the scalp. It is assumed that this would implicate better accuracy in the EEG inverse estimates. However, this does not necessarily hold. The reason for this is that the electrodes create a well conducting shunting "layer" on the scalp which affects the voltage distribution. This may decrease the information obtained and may therefore worsen the inverse estimates. Electrodes in EEG inverse problems are commonly modeled as point electrodes. This model cannot take into account the possible shunting effect of the electrodes. In this study the measurement electrodes are modeled using the so-called complete electrode model which takes into account the actual size of the electrode, the contact impedance between the skin and the electrode and also the shunting effect of the electrodes. In this paper the effects of the electrode size and the contact impedance on the voltage distribution are studied by simulations. It is shown that, depending on the size and the contact impedance of the electrodes, increasing the number of electrodes does not necessarily improve the accuracy of the inverse estimates. We also conclude that the use of the point electrode model is quite adequate in normal EEG studies. The use of a complete electrode model is necessary if electrodes cover more than 50% of the surface area.

Effects of local skull inhomogeneities on EEG source estimation.

The accuracy of the head model affects the solutions of the EEG inverse problems. If a simple three-sphere model and standard conductivity values for brain, skull and scalp regions are used, significant errors may occur in the dipole localisation. One of the most sensitive head model parameters is the conductivity of the skull. A realistic three-dimensional finite-element model provides a method to study the effect of inhomogeneities of the skull on the solutions of EEG inverse problems. In this paper the effect of a local skull conductivity inhomogeneity on source estimation accuracy is analyzed by computer simulations for different numbers of electrodes. It is shown that if the inhomogeneity of the skull conductivity is not taken into account, localisation errors of approximately 1 cm can be encountered in the equivalent current dipole estimation. This modelling error introduces a bias to the solution which cannot be compensated by increasing the number of electrodes.

EMG signal morphology and kinematic parameters in essential tremor and Parkinson's disease patients.

The aim of this work was to differentiate patients with essential tremor from patients with Parkinson's disease. Electromyographic data from biceps brachii muscles and kinematic data from arms during isometric tension of the arms were measured from 17 patients with essential tremor, 35 patients with Parkinson's disease and 40 healthy controls. The EMG signals were divided to smaller segments from which histograms were calculated. The histogram shape was analysed with a feature dimension reduction method, the principal component analysis, and the shape parameters were used to differentiate between different subject groups. Three parameters, RMS-amplitude, sample entropy and peak frequency were determined from the kinematic measurements of the arms. The height and the side differences of the histogram were the most effective for differentiating between essential tremor and Parkinson's disease groups. The histogram parameters of patients with essential tremor were more similar to patients with Parkinson's disease than healthy controls. With this method it was possible to discriminate 13/17 patients with essential tremor from 26/35 patients with Parkinson's disease and 14/17 patients with essential tremor from 29/40 healthy controls. The kinematic parameters of patients with essential tremor were closer to parameters of patients with Parkinson's disease compared to healthy controls. Combining EMG and kinematic analysis did not increase discrimination efficiency but provided more reliability to the discrimination of subject groups.

EMG signal morphology in essential tremor and Parkinson's disease.

The aim of this work was to differentiate patients with essential tremor from patients with Parkinson's disease. The electromyographic signal from the biceps brachii muscle was measured during isometric tension from 17 patients with essential tremor, 35 patients with Parkinson's disease, and 40 healthy controls. The EMG signals were high pass filtered and divided to smaller segments from which histograms were calculated using 200 histogram bins. EMG signal histogram shape was analysed with a feature dimension reduction method, the principal component analysis, and the shape parameters were used to differentiate between different patient groups. The height of the histogram and the side difference between left and right hand were the best discriminators between essential tremor and Parkinson's disease groups. With this method, it was possible to discriminate 13/17 patients with essential tremor from 26/35 patients with Parkinson's disease and 14/17 patients with essential tremor from 29/40 healthy controls.

Linear and nonlinear tremor acceleration characteristics in patients with Parkinson's disease.

The purpose of the study was to evaluate linear and nonlinear tremor characteristics of the hand in patients with Parkinson's disease (PD) and to compare the results with those of healthy old and young control subjects. Furthermore, the aim was to study correlation between tremor characteristics and clinical signs. A variety of nonlinear (sample entropy, cross-sample entropy, recurrence rate, determinism and correlation dimension) and linear (amplitude, spectral peak frequency and total power, and coherence) hand tremor parameters were computed from acceleration measurements for PD patients (n = 30, 68.3 ± 7.8 years), and old (n = 20, 64.2 ± 7.0 years) and young (n = 20, 18.4 ± 1.1 years) control subjects. Nonlinear tremor parameters such as determinism, sample entropy and cross-sample entropy were significantly different between the PD patients and healthy controls. These parameters correlated with the Unified Parkinson's disease rating scale (UPDRS), tremor and finger tapping scores, but not with the rigidity scores. Linear tremor parameters such as the amplitude and the maximum power (power corresponding to peak frequency) also correlated with the clinical findings. No major difference was detected in the tremor characteristics between old and young control subjects. The study revealed that tremor in PD patients is more deterministic and regular when compared to old or young healthy controls. The nonlinear tremor parameters can differentiate patients with PD from healthy control subjects and these parameters may have potential in the assessment of the severity of PD (UPDRS).

Dynamic estimation of cardiac repolarization characteristics during hypoglycemia in healthy and diabetic subjects.

Hypoglycemia is known to affect the repolarization characteristics of the heart, but the mechanisms behind these changes are not completely understood. We analyzed repolarization characteristics continuously from 22 subjects during normoglycemic period, transition period (blood glucose concentration decreasing) and hypoglycemic period from nine healthy controls (Healthy), six otherwise healthy type 1 diabetics (T1DM) and seven type 1 diabetics with disease complications (T1DMc). An advanced principal component regression (PCR)-based method was used for estimating ECG parameters beat-by-beat, and thus, continuous comparison between the repolarization characteristics and blood glucose values was made. We observed that hypoglycemia related ECG changes in the T1DMc group were smaller than changes in the Healthy and T1DM groups. We also noticed that when glucose concentration remained at a low level, the heart rate corrected QT interval prolonged progressively. Finally, a few minutes time lag was observed between the start of hypoglycemia and cardiac repolarization changes. One explanation for these observations could be that hypoglycemia related hormonal changes have a significant role behind the repolarization changes. This could explain at least the observed time lag (hormonal changes are slow) and the lower repolarization changes in the T1DMc group (hormonal secretion lowered in long duration diabetics).

Hypoglycemia detection based on cardiac repolarization features.

Hypoglycemia is known to affect repolarization characteristics of the heart. These changes are shown from ECG by prolonged QT-time and T-wave flattening. In this study we constructed a classifier based on these ECG parameters. By using the classifier we tried to detect hypoglycemic events from measurements of 22 test subjects. Hypoglycemic state was achieved using glucose clamp technique. Used test protocol consisted of three stages: normoglycemic period, transition period (blood glucose concentration decreasing) and hypoglycemic period. Subjects were divided into three groups: 9 healthy controls (Healthy), 6 otherwise healthy type 1 diabetics (T1DM) and 7 type 1 diabetics with disease complications (T1DMc). Detection of hypoglycemic event could be made passably from 15/22 measurements. In addition, we found that detection process is easier for healthy and T1DM groups than T1DMc group diabetics because in T1DMc group subjects' have lower autonomic response to hypoglycemic events. Also we noticed that changes in ECG occurs few minutes after blood glucose is decreased below 3.5 mmol/1.

Analysis of heart rate variability dynamics during propofol and dexmedetomidine anesthesia.

It has been observed that heart rate variability (HRV) diminishes during anesthesia, but the exact mechanisms causing it are not completely understood. The aim of this paper was to study the dynamics of HRV during low dose propofol (N=9) and dexmedetomidine (N=8) anesthesia by using state-of-the-art time-varying methods, and thereby ultimately try to improve the safety of anesthesia. The time-varying spectrum is estimated by using a Kalman smoother approach. The results show that there is an overall increase in HRV and decrease in heart rate prior to loss of consciousness. For dexmedetomidine these changes are more considerable than for propofol. For dexmedetomidine the variability also seems to start decreasing right after loss of consciousness, whereas for propofol HRV continues increasing.

Novel parameters of surface EMG in patients with Parkinson's disease and healthy young and old controls.

The aim of this study was to evaluate a variety of traditional and novel surface electromyography (SEMG) characteristics of biceps brachii muscle in patients with Parkinson's disease (PD) and compare the results with the healthy old and young control subjects. Furthermore, the aim was to define the optimal biceps brachii loading level that would most likely differentiate patients from controls. The results indicated that such nonlinear SEMG parameters as %Recurrence, %Determinism and SEMG distribution kurtosis, correlation dimension and sample entropy were significantly different between the PD patients and healthy controls. These novel nonlinear parameters, unlike traditional spectral or amplitude parameters, correlated with the Unified Parkinson's Disease Rating Scale (UPDRS) and finger tapping scores. The most significant between group differences were found in the loading condition where no additional weights were applied in isometric elbow flexion. No major difference of SEMG characteristics was detected between old and young control subjects. In conclusion, the novel SEMG parameters can differentiate the patients with PD from healthy control subjects and these parameters may have potential in the assessment of the severity of PD.

Time-varying spectrum estimation of heart rate variability signals with Kalman smoother algorithm.

A time-varying parametric spectrum estimation method for analyzing dynamics of heart rate variability (HRV) signals is presented. In the method, HRV signal is first modeled with a time-varying autoregressive model and the model parameters are solved recursively with a Kalman smoother algorithm. Time-varying spectrum estimates are then obtained from the estimated model parameters. The obtained spectrum can be further decomposed into separate components, which is especially advantageous in HRV applications where low frequency (LF) and high frequency (HF) components are generally aimed to be distinguished. As case studies, the dynamics of HRV signals recorded during 1) orthostatic test, 2) exercise test and 3) simulated driving task are analyzed.