Aydin Akan

Decreased global field synchronization of multichannel frontal EEG measurements in obsessive-compulsive disorders

Global field synchronization (GFS) quantifies the synchronization level of brain oscillations. The GFS method has been introduced to measure functional synchronization of EEG data in the frequency domain. GFS also detects phase interactions between EEG signals acquired from all of the electrodes. If a considerable amount of local brain neurons has the same phase, these neurons appear to interact with each other. EEG data were received from 17 obsessive-compulsive disorder (OCD) patients and 17 healthy controls (HC). OCD effects on local and large-scale brain circuits were studied. Analysis of the GFS results showed significantly decreased values in the delta and full frequency bands. This research suggests that OCD causes synchronization disconnection in both the frontal and large-scale regions. This may be related to motivational, emotional and cognitive dysfunctions.

Cyclostationary Analysis of Respiratory Signals with Application of Rate Determination

Respiratory signals are periodic-like signals where the noisy periodic pattern repeats itself. Therefore, based on a stationarity assumption, autocorrelation function contains noisy cycles in time-lag with the same rate as the respiration rate. In this work, cyclostationarity test is performed on the respiratory signals in order to determine cyclic characteristics of the time varying autocorrelation function. Our specific aim is to check whether the cycle period in time corresponds to the respiration rate. Lung simulator was used to generate the respiratory signals. Time varying autocorrelation variance was computed by using both the modified windowed, and the blocked signal methods. Our simulations resulted that the cycle period was the same as the respiration period. Moreover, we observed that cyclic frequencies corresponded to the respiratory rate and its harmonics.

Seizure onset detection based on frequency domain metric of empirical mode decomposition

This paper explores the data-driven properties of the empirical mode decomposition (EMD) for detection of epileptic seizures. A new method in frequency domain is presented to analyze intrinsic mode functions (IMFs) decomposed by EMD. They are used to determine whether the electroencephalogram (EEG) recordings contain seizure or not. Energy levels of the IMFs are extracted as threshold level to detect the changes caused by seizure activity. A scalar value energy resulting from the energy levels is individually used as an indicator of the epileptic EEG without the requirements of multidimensional feature vector and complex machine learning algorithms. The proposed methods are tested on different EEG recordings to evaluate the effectiveness of the proposed method and yield accuracy rate up to 97.89%.

Analysis of gastric myoelectrical activity from the electrogastrogram signals based on wavelet transform and line length feature

Electrogastrogram is used for the abdominal surface measurement of the gastric electrical activity of the human stomach. The electrogastrogram technique has significant value as a clinical tool because careful electrogastrogram signal recordings and analyses play a major role in determining the propagation and coordination of gastric myoelectric abnormalities. The aim of this article is to evaluate electrogastrogram features calculated by line length features based on the discrete wavelet transform method to differentiate healthy control subjects from patients with functional dyspepsia and diabetic gastroparesis. For this analysis, the discrete wavelet transform method was used to extract electrogastrogram signal characteristics. Next, line length features were calculated for each sub-signal, which reflect the waveform dimensionality variations and represent a measure of sensitivity to differences in signal amplitude and frequency. The analysis was carried out using a statistical analysis of variance test. The results obtained from the line length analysis of the electrogastrogram signal prove that there are significant differences among the functional dyspepsia, diabetic gastroparesis, and control groups. The electrogastrogram signals of the control subjects had a significantly higher line length than those of the functional dyspepsia and diabetic gastroparesis patients. In conclusion, this article provides new methods with increased accuracy obtained from electrogastrogram signal analysis. The electrogastrography is an effective and non-stationary method to differentiate diabetic gastroparesis and functional dyspepsia patients from the control group. The proposed method can be considered a key test and an essential computer-aided diagnostic tool for detecting gastric myoelectric abnormalities in diabetic gastroparesis and functional dyspepsia patients.

Emotion recognition based on time–frequency distribution of EEG signals using multivariate synchrosqueezing transform

Abstract This paper investigates the feasibility of using time–frequency (TF) representation of EEG signals for emotional state recognition. A recent and advanced TF analyzing method, multivariate synchrosqueezing transform (MSST) is adopted as a feature extraction method due to multi-channel signal processing and compact component localization capabilities. First, the 32 participants EEG recordings from DEAP emotional EEG database are analyzed using MSST to reveal oscillations. Second, independent component analysis (ICA), and feature selection are applied to reduce the high dimensional 2D TF distribution without losing distinctive component information in the 2D image. Thus, only one method for feature extraction using MSST is performed to analyze time, and frequency-domain properties of the EEG signals instead of using some signal analyzing combinations (e.g., power spectral density, energy in bands, Hjorth parameters, statistical values, and time differences etc.). As well, the TF-domain reduction performance of ICA is compared to non-negative matrix factorization (NMF) to discuss the accuracy levels of high/low arousal, and high/low valence emotional state recognition. The proposed MSST-ICA feature extraction approach yields up to correct rates of 82.11%, and 82.03% for arousal, and valence state recognition using artificial neural network. The performances of the MSST and ICA are compared with Wigner-Ville distribution (WVD) and NMF to investigate the effects of TF distributions as feature set with reduction techniques on emotion recognition.

Ensemble learning approaches to classification of pulmonary nodules

Lung cancer is one of the primary causes of cancer-related death worldwide. A computer-aided detection (CAD) can help radiologists by offering a second opinion and making the whole process faster at an early level. In this study, we propose a new classification approach for pulmonary nodule detection from CT imagery by using morphological features of nodule patterns. Ensemble learning approaches are used for classification process and overall detection performance is evaluated. Results are compared to similar techniques in the literature by using standard measures. The performance of the proposed system with random forest based on ensemble learning approaches results in an overall accuracy of 98.7 % with a sensitivity of 100 % and specificity of 97.3 % in training data set and an overall accuracy of 80.7 % with a sensitivity of 80.7 % and specificity of 80.6 % in testing dataset.