A New Hybrid Approach Based on Time Frequency Images and Deep Learning Methods for Diagnosis of Migraine Disease and Investigation of Stimulus Effect.

Migraine is a neurological disorder that is associated with severe headaches and seriously affects the lives of patients. Diagnosing Migraine Disease (MD) can be laborious and time-consuming for specialists. For this reason, systems that can assist specialists in the early diagnosis of MD are important. Although migraine is one of the most common neurological diseases, there are very few studies on the diagnosis of MD, especially electroencephalogram (EEG)-and deep learning (DL)-based studies. For this reason, in this study, a new system has been proposed for the early diagnosis of EEG- and DL-based MD. In the proposed study, EEG signals obtained from the resting state (R), visual stimulus (V), and auditory stimulus (A) from 18 migraine patients and 21 healthy control (HC) groups were used. By applying continuous wavelet transform (CWT) and short-time Fourier transform (STFT) methods to these EEG signals, scalogram-spectrogram images were obtained in the time-frequency (T-F) plane. Then, these images were applied as inputs in three different convolutional neural networks (CNN) architectures (AlexNet, ResNet50, SqueezeNet) that proposed deep convolutional neural network (DCNN) models and classification was performed. The results of the classification process were evaluated, taking into account accuracy (acc.), sensitivity (sens.), specificity (spec.), and performance criteria, and the performances of the preferred methods and models in this study were compared. In this way, the situation, method, and model that showed the most successful performance for the early diagnosis of MD were determined. Although the classification results are close to each other, the resting state, CWT method, and AlexNet classifier showed the most successful performance (Acc: 99.74%, Sens: 99.9%, Spec: 99.52%). We think that the results obtained in this study are promising for the early diagnosis of MD and can be of help to experts.

Quantitative Electroencephalography Analysis for Improved Assessment of Consciousness Levels in Deep Coma Patients Using a Proposed Stimulus Stage.

"Coma" is defined as an inability to obey commands, to speak, or to open the eyes. So, a coma is a state of unarousable unconsciousness. In a clinical setting, the ability to respond to a command is often used to infer consciousness. Evaluation of the patient's level of consciousness (LeOC) is important for neurological evaluation. The Glasgow Coma Scale (GCS) is the most widely used and popular scoring system for neurological evaluation and is used to assess a patient's level of consciousness. The aim of this study is the evaluation of GCSs with an objective approach based on numerical results. So, EEG signals were recorded from 39 patients in a coma state with a new procedure proposed by us in a deep coma state (GCS: between 3 and 8). The EEG signals were divided into four sub-bands as alpha, beta, delta, and theta, and their power spectral density was calculated. As a result of power spectral analysis, 10 different features were extracted from EEG signals in the time and frequency domains. The features were statistically analyzed to differentiate the different LeOC and to relate with the GCS. Additionally, some machine learning algorithms have been used to measure the performance of the features for distinguishing patients with different GCSs in a deep coma. This study demonstrated that GCS 3 and GCS 8 patients were classified from other levels of consciousness in terms of decreased theta activity. To the best of our knowledge, this is the first study to classify patients in a deep coma (GCS between 3 and 8) with 96.44% classification performance.

A native extracellular matrix material for tissue engineering applications: Characterization of pericardial fluid.

Tissue engineering applications are widely used to repair and regenerate damaged tissues and organs. A scaffold, which is an important component in tissue engineering, provides a 3D environment for cells. In this study, the usability of PF components for the production of an ideal scaffold was investigated. For this aim, pericardial fluid (PF) was harvested from the bovine heart, then its structure and components were characterized. The results of Raman spectroscopy analysis, histological staining, and scanning electron microscopy (SEM) shows that the pericardial fluid contains collagen type I and IV, elastin, fibrin, and glycosaminoglycan (GAG), which are native extracellular matrix (ECM) components. The results demonstrated that (i) PF contains native ECM proteins and GAG such as collagen types I, III, and IV, elastin, and fibrin. (ii) The PF is highly similar to the native ECM structure. (iii) PF can significantly contribute to many tissue engineering studies as a native ECM material to increase the biocompatibility of biomaterials and to several in vitro/in vivo cell culture studies. (iv) PF containing multiple ECM molecules, can be used alone or together with hyaluronic acid, poly(ethylene glycol) (PEG), alginate, chitosan, matrigel, and gelatin methacryloyl (GelMA) materials in bioprinting systems for eliminating the disadvantages of these materials.

Distinguishing Parkinson's Disease with GLCM Features from the Hankelization of EEG Signals.

This study proposes a novel method that uses electroencephalography (EEG) signals to classify Parkinson's Disease (PD) and demographically matched healthy control groups. The method utilizes the reduced beta activity and amplitude decrease in EEG signals that are associated with PD. The study involved 61 PD patients and 61 demographically matched controls groups, and EEG signals were recorded in various conditions (eyes closed, eyes open, eyes both open and closed, on-drug, off-drug) from three publicly available EEG data sources (New Mexico, Iowa, and Turku). The preprocessed EEG signals were classified using features obtained from gray-level co-occurrence matrix (GLCM) features through the Hankelization of EEG signals. The performance of classifiers with these novel features was evaluated using extensive cross-validations (CV) and leave-one-out cross-validation (LOOCV) schemes. This method under 10 × 10 fold CV, the method was able to differentiate PD groups from healthy control groups using a support vector machine (SVM) with an accuracy of 92.4 ± 0.01, 85.7 ± 0.02, and 77.1 ± 0.06 for New Mexico, Iowa, and Turku datasets, respectively. After a head-to-head comparison with state-of-the-art methods, this study showed an increase in the classification of PD and controls.

Detection of ADHD from EEG signals using new hybrid decomposition and deep learning techniques.

Objective.Attention deficit hyperactivity disorder (ADHD) is considered one of the most common psychiatric disorders in childhood. The incidence of this disease in the community draws an increasing graph from the past to the present. While the ADHD diagnosis is basically made with the psychiatric tests, there is no active clinically used objective diagnostic tool. However, some studies in the literature has reported development of an objective diagnostic tool that facilitates the diagnosis of ADHD.Approach.In this study, it was aimed to develop an objective diagnostic tool for ADHD using electroencephalography (EEG) signals. In the proposed method, EEG signals were decomposed into subbands by robust local mode decomposition and variational mode decomposition techniques. These subbands and the EEG signals were fed as input data to the deep learning algorithm designed in the study.Main results.As a result, an algorithm has been put forward that distinguishes over 95% of ADHD and healthy individuals through using a 19-channel EEG signal. In addition, a classification accuracy of over 87% was obtained by the proposed approach of EEG signal decomposition followed by data processing in the designed deep learning algorithm.Significance.The findings of the current research enrich the literature based on originality and proposed method can be used as a clinical diagnostic tool in the near future.

Using electrooculography with visual stimulus tracking test in diagnosing of ADHD: findings from machine learning algorithms.

BACKGROUND: Attention deficit hyperactivity disorder (ADHD), one of the most common neurodevelopmental disorders in childhood, is diagnosed clinically by assessing the symptoms of inattention, hyperactivity, and impulsivity. Also, there are limited objective assessment tools to support the diagnosis. Thus, in this study, a new electrooculography (EOG) based on visual stimulus tracking to support the diagnosis of ADHD was proposed. METHODS: Reference stimulus one-to-one tracking numbers (RSOT) and colour game detection (CGD) were applied to 53 medication-free children with ADHD and 36 healthy controls (HCs). Also, the test was applied six months after the treatment to children with ADHD. Parameters obtained during the visual stimulus tracking test were analyzed and Higuchi fractal dimension (HFD) and Hjorth parameters were calculated for all EOG records. RESULTS: The average test success rate was higher in HCs than in children with ADHD. Based on machine learning algorithms, the proposed system can distinguish drug-free ADHD patients from HCs with an 89.13% classification performance and also distinguish drug-free children from treated children with an 80.47% classification performance. DISCUSSION: The findings showed that the proposed system could be helpful to support the diagnosis of ADHD and the follow-up of the treatment.

A novel approach for detection of dyslexia using convolutional neural network with EOG signals.

Dyslexia is a learning disability in acquiring reading skills, even though the individual has the appropriate learning opportunity, adequate education, and appropriate sociocultural environment. Dyslexia negatively affects children's educational development; hence, early detection is highly important. Electrooculogram (EOG) signals are one of the most frequently used physiological signals in human-computer interfaces applications. EOG is a method based on the examination of the electrical potential of eye movements. The advantages of EOG-based systems are non-invasive, affordable, easy to record, and can be processed in real time. In this paper, a novel 1D CNN approach using EOG signals is proposed for the diagnosis of dyslexia. The proposed approach aims to diagnose dyslexia using EOG signals that are recorded simultaneously during reading texts, which are prepared in different typefaces and fonts. EOG signals were recorded from both horizontal and vertical channels, thus comparing the success of vertical and horizontal EOG signals in detecting dyslexia. The proposed approach provided an effective classification without requiring any hand-crafted feature extraction techniques. The proposed method achieved classifier accuracy of 98.70% and 80.94% for horizontal and vertical channel EOG signals, respectively. The results show that the EOG signals-based approach gives successful results for the diagnosis of dyslexia.

Classification of Depth of Coma Using Complexity Measures and Nonlinear Features of Electroencephalogram Signals.

In recent years, some electrophysiological analysis methods of consciousness have been proposed. Most of these studies are based on visual interpretation or statistical analysis, and there is hardly any work classifying the level of consciousness in a deep coma. In this study, we perform an analysis of electroencephalography complexity measures by quantifying features efficiency in differentiating patients in different consciousness levels. Several measures of complexity have been proposed to quantify the complexity of signals. Our aim is to lay the foundation of a system that will objectively define the level of consciousness by performing a complexity analysis of Electroencephalogram (EEG) signals. Therefore, a nonlinear analysis of EEG signals obtained with a recording scheme proposed by us from 39 patients with Glasgow Coma Scale (GCS) between 3 and 8 was performed. Various entropy values (approximate entropy, permutation entropy, etc.) obtained from different algorithms, Hjorth parameters, Lempel-Ziv complexity and Kolmogorov complexity values were extracted from the signals as features. The features were analyzed statistically and the success of features in classifying different levels of consciousness was measured by various classifiers. Consequently, levels of consciousness in deep coma (GCS between 3 and 8) were classified with an accuracy of 90.3%. To the authors' best knowledge, this is the first demonstration of the discriminative nonlinear features extracted from tactile and auditory stimuli EEG signals in distinguishing different GCSs of comatose patients.

Detection of amyotrophic lateral sclerosis disease from event-related potentials using variational mode decomposition method.

This study, it was aimed to contribute to the literature on Amyotrophic lateral sclerosis (ALS) diagnosis and Brain-Computer Interface (BCI) technologies by analyzing the electroencephalography (EEG) signals obtained as a result of visual stimuli and attention from ALS patients and healthy controls. It was observed that the success rate significantly increased both in the occipital and central regions in all classifiers, especially in the entropy features. The most successful classification was obtained with the Naïve Bayes (NB) classifier using the Morphological Features (MF) + Variational Mode Decomposition (VMD) -Entropy features at 88.89% in the occipital region and 94.44% in the central region.

Analysis of Consciousness Level Using Galvanic Skin Response during Therapeutic Effect.

The neurological status of patients in the Intensive Care Units (ICU) is determined by the Glasgow Coma Scale (GCS). Patients in coma are thought to be unaware of what is happening around them. However, many studies show that the family plays an important role in the recovery of the patient and is a great emotional resource. In this study, Galvanic Skin Response (GSR) signals were analyzed from 31 patients with low consciousness levels between GCS 3 and 8 to determine relationship between consciousness level and GSR signals as a new approach. The effect of family and nurse on unconscious patients was investigated by GSR signals recorded with a new proposed protocol. The signals were recorded during conversation and touching of the patient by the nurse and their families. According to numerical results, the level of consciousness can be separated using GSR signals. Also, it was found that family and nurse had statistically significant effects on the patient. Patients with GCS 3,4, and 5 were considered to have low level of consciousness, while patients with GCS 6,7, and 8 were considered to have high level of consciousness. According to our results, it is obtained lower GSR amplitude in low GCS (3, 4, 5) compared to high GCS (7, 8). It was concluded that these patients were aware of therapeutic affect although they were unconscious. During the classification stage of this study, the class imbalance problem, which is common in medical diagnosis, was solved using Synthetic Minority Over-Sampling Technique (SMOTE), Adaptive Synthetic Sampling (ADASYN) and random oversampling methods. In addition, level of consciousness was classified with 92.7% success using various decision tree algorithms. Random Forest was the method which provides higher accuracy compared to all other methods. The obtained results showed that GSR signal analysis recorded in different stages gives very successful GCS score classification performance according to literature studies.

Diagnosis of attention-deficit hyperactivity disorder using EOG signals: a new approach.

A system based on objective data was developed in the diagnosis and follow-up of attention-deficit hyperactivity disorder (ADHD) in this study. First of all, an electronic circuit, with a two-channel instrumentation amplifier designed to detect eye movements in the horizontal and vertical directions via surface electrodes, was developed to obtain the electrooculogram (EOG) signals. In order to provide a controlled analysis of eye movements during the reception of the signal, an attention test with visual stimulus software was developed. Eight patients with ADHD and eight healthy subjects were asked to monitor the stimulus images on the screen in the reference directions of the test system while recording EOG signals. According to the results of the t-test, no significant difference was found (p=0.11) between the healthy group and the reference movement information, whereas a significant difference was found between patients and the reference motion information (p=0.049). According to these results, it was seen that the number of eye movements of healthy individuals was statistically significant. In addition, they were inconsistent with the reference movement information. The level of significance was found to be low in patients. In this study, a new method is presented to test and diagnose individuals who were attention deficit.

Analysis of heart rate variability during auditory stimulation periods in patients with schizophrenia.

The vulnerability-stress model is a hypothesis for symptom development in schizophrenia patients who are generally characterized by cardiac autonomic dysfunction. Therefore, measures of heart rate variability (HRV) have been widely used in schizophrenics for assessing altered cardiac autonomic regulations. The goal of this study was to analyze HRV of schizophrenia patients and healthy control subjects with exposure to auditory stimuli. More specifically, this study examines whether schizophrenia patients may exhibit distinctive time and frequency domain parameters of HRV from control subjects during at rest and auditory stimulation periods. Photoplethysmographic signals were used in the analysis of HRV. Nineteen schizophrenic patients and twenty healthy control subjects were examined during rest periods, while exposed to periods of white noise (WN) and relaxing music. Results indicate that HRV in patients was lower than that of control subjects indicating autonomic dysfunction throughout the entire experiment. In comparison with control subjects, patients with schizophrenia exhibited lower high-frequency power and a higher low-frequency to high-frequency ratio. Moreover, while WN stimulus decreased parasympathetic activity in healthy subjects, no significant changes in heart rate and frequency-domain HRV parameters were observed between the auditory stimulation and rest periods in schizophrenia patients. We can conclude that HRV can be used as a sensitive index of emotion-related sympathetic activity in schizophrenia patients.

A novel approach to speckle noise filtering based on Artificial Bee Colony algorithm: an ultrasound image application.

In this study a novel approach based on 2D FIR filters is presented for denoising digital images. In this approach the filter coefficients of 2D FIR filters were optimized using the Artificial Bee Colony (ABC) algorithm. To obtain the best filter design, the filter coefficients were tested with different numbers (3×3, 5×5, 7×7, 11×11) and connection types (cascade and parallel) during optimization. First, the speckle noise with variances of 1, 0.6, 0.8 and 0.2 respectively was added to the synthetic test image. Later, these noisy images were denoised with both the proposed approach and other well-known filter types such as Gaussian, mean and average filters. For image quality determination metrics such as mean square error (MSE), peak signal-to-noise ratio (PSNR) and signal-to-noise ratio (SNR) were used. Even in the case of noise having maximum variance (the most noisy), the proposed approach performed better than other filtering methods did on the noisy test images. In addition to test images, speckle noise with a variance of 1 was added to a fetal ultrasound image, and this noisy image was denoised with very high PSNR and SNR values. The performance of the proposed approach was also tested on several clinical ultrasound images such as those obtained from ovarian, abdomen and liver tissues. The results of this study showed that the 2D FIR filters designed based on ABC optimization can eliminate speckle noise quite well on noise added test images and intrinsically noisy ultrasound images.

Non-invasive diagnosis of stress urinary incontinence sub types using wavelet analysis, shannon entropy and principal component analysis.

Urinary incontinence is a common female disorder. Although generally not a serious condition, it negatively affects the lifestyle and daily activity of subjects. Stress urinary incontinence (SUI) is the most versatile of several incontinence types and is distinguished by physical degeneration of the continence-providing mechanism. Some surgical treatment methods exist, but the success of the surgery mainly depends upon a correct diagnosis. Diagnosis has two major steps: subjects who are suffering from true SUI must be identified, and the SUI sub-type must be determined, because each sub-type is treated with a different surgery. The first step is straightforward and uses standard identification methods. The second step, however, requires invasive, uncomfortable urodynamic studies that are difficult to apply. Many subjects try to cope with the disorder rather than seek treatment from health care providers, in part because of the invasive diagnostic methods. In this study, a diagnostic method with a success rate comparable to that of urodynamic studies is presented. This new method has some advantages over the current one. First, it is noninvasive; data are collected using Doppler ultrasound recording. Second, it requires no special tools and is easy to apply, relatively inexpensive, faster and more hygienic.

Comparison of short-time Fourier transform and Eigenvector MUSIC methods using discrete wavelet transform for diagnosis of atherosclerosis.

In this paper, a more effective use of Doppler techniques is presented for the purpose of diagnosing atherosclerosis in its early stages using the carotid artery Doppler signals. The power spectral density (PSD) graphics are obtained by applying the short-time Fourier transform (STFT)-Welch and the Eigenvector MUSIC methods to the discrete wavelet transform (DWT) of Doppler signals. The PSDs for the fourth approximation component (A4) of both methods estimated that the patients with atherosclerosis in its early phase had lower maximum frequency components. On the other hand, the healthy subjects had higher maximum frequency components. The area under the curve (AUC), which belongs to the receiver operating characteristic (ROC) curve for the frequency level of the maximum PSDs of the A4 approximation obtained from the STFT modeling, is computed as 0.97. The AUC for the MUSIC modeling is computed as 0.996. The AUC belonging to the ROC curve for the higher maximum frequency component is computed as 0.87. The AUC belonging to the ROC curve for the test parameter of the frequency level of the maximum PSDs derived from the MUSIC modeling is determined to be 0.882. The results of this study clearly demonstrate that it is possible to distinguish between the healthy people and the patients with atherosclerosis by using the frequency level of the maximum PSDs for the A4 approximation. Furthermore, it is concluded that the power of Eigenvector-MUSIC method in terms of the resolution of the high frequencies is better than that of the STFT methods.

Use of kernel functions in artificial immune systems for the nonlinear classification problems.

Due to the fact that there exist only a small number of complex systems in artificial immune systems (AISs) that solve nonlinear problems, there is a need to develop nonlinear AIS approaches that would be among the well-known solution methods. In this study, we developed a kernel-based AIS to compensate for this deficiency by providing a nonlinear structure via transformation of distance calculations in the clonal selection models of classical AIS to kernel space. Applications of the developed system were conducted on Statlog heart disease dataset, which was taken from the University of California, Irvine Machine-Learning Repository, and on Doppler sonograms to diagnose atherosclerosis disease. The system obtained a classification accuracy of 85.93% for the Statlog heart disease dataset, while it achieved a 99.09% classification success for the Doppler dataset. With these results, our system seems to be a potential solution method, and it may be considered as a suitable method for hard nonlinear classification problems.

Medical diagnosis of atherosclerosis from Carotid Artery Doppler Signals using principal component analysis (PCA), k-NN based weighting pre-processing and Artificial Immune Recognition System (AIRS).

In this study, we proposed a new medical diagnosis system based on principal component analysis (PCA), k-NN based weighting pre-processing, and Artificial Immune Recognition System (AIRS) for diagnosis of atherosclerosis from Carotid Artery Doppler Signals. The suggested system consists of four stages. First, in the feature extraction stage, we have obtained the features related with atherosclerosis disease using Fast Fourier Transformation (FFT) modeling and by calculating of maximum frequency envelope of sonograms. Second, in the dimensionality reduction stage, the 61 features of atherosclerosis disease have been reduced to 4 features using PCA. Third, in the pre-processing stage, we have weighted these 4 features using different values of k in a new weighting scheme based on k-NN based weighting pre-processing. Finally, in the classification stage, AIRS classifier has been used to classify subjects as healthy or having atherosclerosis. Hundred percent of classification accuracy has been obtained by the proposed system using 10-fold cross validation. This success shows that the proposed system is a robust and effective system in diagnosis of atherosclerosis disease.

Usage of a novel, similarity-based weighting method to diagnose atherosclerosis from carotid artery Doppler signals.

In this paper, we have proposed a novel similarity-based weighting method (SBWM), which combines similarity measure and weighting based on trend association (WBTA) method proposed by Sun Yi et al. (ICNN&B international conference, vol 1, pp 266-269, 2005). The aim of this study is to improve the classification accuracy of atherosclerosis, which is a common disease among the public. The proposed method consists of three parts: (1) feature extraction part related with atherosclerosis disease using fast Fourier transformation (FFT) modeling and calculation of maximum frequency envelope of sonograms, (2) data pre-processing part using SBWM, including different similarity measures such as cosine amplitude method, max-min method, absolute exponential method, and exponential similarity coefficient, and (3) classification part using artificial immune recognition system (AIRS) and Fuzzy-AIRS classifier algorithms. While AIRS and Fuzzy-AIRS algorithms obtained 71.92 and 78.94% success rates, respectively, the combination of SBWM with classifier algorithms including AIRS and Fuzzy-AIRS obtained 100% success rate on all the similarity measures. These results show that SBWM has produced very promising results in the classification of atherosclerosis from carotid artery Doppler signals. In future, we will use a larger dataset to test the proposed method.

A new supervised classification algorithm in artificial immune systems with its application to carotid artery Doppler signals to diagnose atherosclerosis.

Because of its self-regulating nature, immune system has been an inspiration source for usually unsupervised learning methods in classification applications of Artificial Immune Systems (AIS). But classification with supervision can bring some advantages to AIS like other classification systems. Indeed, there have been some studies, which have obtained reasonable results and include supervision in this branch of AIS. In this study, we have proposed a new supervised AIS named as Supervised Affinity Maturation Algorithm (SAMA) and have presented its performance results through applying it to diagnose atherosclerosis using carotid artery Doppler signals as a real-world medical classification problem. We have employed the maximum envelope of the carotid artery Doppler sonograms derived from Autoregressive (AR) method as an input of proposed classification system and reached a maximum average classification accuracy of 98.93% with 10-fold cross-validation method used in training-test portioning. To evaluate this result, comparison was done with Artificial Neural Networks and Decision Trees. Our system was found to be comparable with those systems, which are used effectively in literature with respect to classification accuracy and classification time. Effects of system's parameters were also analyzed in performance evaluation applications. With this study and other possible contributions to AIS, classification algorithms with effective performances can be developed and potential of AIS in classification can be further revealed.

Determining fractal dimension of umbilical artery Doppler signals using Hurst exponent.

Doppler signals from the umbilical artery of 20 women with normal pregnancy between 18 and 20 weeks of gestation were recorded. The AR spectral analysis method has been used to obtain the Doppler sonograms of umbilical artery belonging to normal pregnant subjects and fractal dimension curves were calculated using Hurst exponent. RI; PI and S/D indexes have been calculated from the maximum frequency envelope of Doppler sonograms and from the fractal dimension curve. Area under the curve from ROC curve for RI, PI and S/D indexes derived from maximum frequency waveform were calculated as 0.931, 0.959, 0.938, respectively and area under the curve for RI, PI and S/D indexes derived from fractal dimension curve were calculated as 0.933, 0.961, and 0.941, respectively. These results show that, the Doppler indexes derived from fractal dimension curve are as sensitive as Doppler indexes derived from maximum velocity curve. Power Spectral Density graphics were derived from Doppler signals and Hurst exponent values calculated to evaluate the blood flow changing during pregnancy. ROC curve for PSD(HURST) index was calculated as 0.97. According to this result, PSD(HURST) index is more sensitive to detect the blood flow changing than traditional Doppler indexes.