Machine Learning: Science and Technology, Jan 17, 2024
Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years ... more Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years thanks to their potential to enhance rehabilitation and control of prosthetic devices for individuals with motor disabilities. However, accurate classification of motor imagery signals remains a challenging task due to the high inter-subject variability and non-stationarity in the electroencephalogram (EEG) data. In the context of MI-BCIs, with limited data availability, the acquisition of EEG data can be difficult. In this study, several data augmentation techniques have been compared with the proposed data augmentation technique adaptive cross-subject segment replacement (ACSSR). This technique, in conjunction with the proposed deep learning framework, allows for a combination of similar subject pairs to take advantage of one another and boost the classification performance of MI-BCIs. The proposed framework features a multi-domain feature extractor based on common spatial patterns with a sliding window and a parallel two-branch convolutional neural network. The performance of the proposed methodology has been evaluated on the multi-class BCI Competition IV Dataset 2a through repeated 10-fold cross-validation. Experimental results indicated that the implementation of the ACSSR method (80.47%) in the proposed framework has led to a considerable improvement in the classification performance compared to the classification without data augmentation (77.63%), and other fundamental data augmentation techniques used in the literature. The study contributes to the advancements for the development of effective MI-BCIs by showcasing the ability of the ACSSR method to address the challenges in motor imagery signal classification tasks.
Attention-deficit/hyperactivity disorder (ADHD) is the most common neurodevelopmental disorder an... more Attention-deficit/hyperactivity disorder (ADHD) is the most common neurodevelopmental disorder and is characterized by symptoms of inattention and/or hyperactivity and impulsivity. In the current study, we obtained quantitative EEG (QEEG) recordings of 51 children aged between 6 and 12 years before the initiation of methylphenidate treatment. The relationship between changes in the scores of ADHD symptoms and initial QEEG features (power/power ratios values) were assessed. In addition, the children were classified as responder and nonresponder according to the ratio of their response to the medication (>25% improvement after medication). Logistic regression analyses were performed to analyze the accuracy of QEEG features for predicting responders. The findings indicate that patients with increased delta power at F8, theta power at Fz, F4, C3, Cz, T5, and gamma power at T6 and decreased beta powers at F8 and P3 showed more improvement in ADHD hyperactivity symptoms. In addition, i...
Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years ... more Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years thanks to their potential to enhance rehabilitation and control of prosthetic devices for individuals with motor disabilities. However, accurate classification of motor imagery signals remains a challenging task due to the high inter-subject variability and non-stationarity in the electroencephalogram (EEG) data. In the context of MI-BCIs, with limited data availability, the acquisition of EEG data can be difficult. In this study, several data augmentation techniques have been compared with the proposed data augmentation technique adaptive cross-subject segment replacement (ACSSR). This technique, in conjunction with the proposed deep learning framework, allows for a combination of similar subject pairs to take advantage of one another and boost the classification performance of MI-BCIs. The proposed framework features a multi-domain feature extractor based on common spatial patterns with a sliding window and a parallel two-branch convolutional neural network. The performance of the proposed methodology has been evaluated on the multi-class BCI Competition IV Dataset 2a through repeated 10-fold cross-validation. Experimental results indicated that the implementation of the ACSSR method (80.47%) in the proposed framework has led to a considerable improvement in the classification performance compared to the classification without data augmentation (77.63%), and other fundamental data augmentation techniques used in the literature. The study contributes to the advancements for the development of effective MI-BCIs by showcasing the ability of the ACSSR method to address the challenges in motor imagery signal classification tasks.
Objective. In the last decades, machine learning approaches have been widely used to distinguish ... more Objective. In the last decades, machine learning approaches have been widely used to distinguish Parkinson’s disease (PD) and many other neuropsychiatric diseases. They also speed up the clinicians and facilitate decision-making for several conditions with similar clinical symptoms. The current study attempts to detect PD with dementia (PDD) by event-related oscillations (EROs) during cognitive processing in two modalities, i.e. auditory and visual. Approach. The study was conducted to discriminate PDD from healthy controls (HC) using event-related phase-locking factors in slow frequency ranges (delta and theta) during visual and auditory cognitive tasks. Seventeen PDD and nineteen HC were included in the study, and linear discriminant analysis was used as a classifier. During classification analysis, multiple settings were implemented by using different sets of channels (overall, fronto-central and temporo-parieto-occipital (TPO) region), frequency bands (delta-theta combined, delta, theta, and low theta), and time of interests (0.1–0.7 s, 0.1–0.5 s and 0.1–0.3 s for delta, delta-theta combined; 0.1–0.4 s for theta and low theta) for spatial-spectral-temporal searchlight procedure. Main results. The classification performance results of the current study revealed that if visual stimuli are applied to PDD, the delta and theta phase-locking factor over fronto-central region have a remarkable contribution to detecting the disease, whereas if auditory stimuli are applied, the phase-locking factor in low theta over TPO and in a wider range of frequency (1–7 Hz) over the fronto-central region classify HC and PDD with better performances. Significance. These findings show that the delta and theta phase-locking factor of EROs during visual and auditory stimuli has valuable contributions to detecting PDD.
Lack of insight is a neurocognitive problem commonly encountered in patients with psychotic disor... more Lack of insight is a neurocognitive problem commonly encountered in patients with psychotic disorders that negatively affects treatment compliance and prognosis. Measurement of insight is based on self-report scales, which are limited due to subjectivity. This study aimed to determine the correlation between resting state beta and gamma power in 23 patients with schizophrenia and insight. It was observed that as beta and gamma power measured via qualitative electroencephalography (qEEG) increased the level of insight decreased. Negative correlation was found in F3, C3, Cz for gamma activity and in F3 and C3 for beta activity. This finding indicates that resting state qEEG could be used to evaluate the level of insight in patients with schizophrenia.
Unexpected events in the environment elicit the orienting response that protects humans from dang... more Unexpected events in the environment elicit the orienting response that protects humans from dangerous situations and there is great importance in identifying these events, especially in aging. The aims of the current study are attempting to find which classification model exhibits the best performance by means of event-related spectral perturbation (ERSP) features based on EEG and to understand which frequency bands, and time windows, contribute most to the classification of external stimuli. The data of 20 healthy elderly participants were included in the study and the 3-Stimulation auditory oddball paradigm was applied to participants. Different classifiers including Support Vector Machine (SVM) with Linear and Polynomial kernels, Linear Discriminant Analysis (LDA), and Naive Bayes were fed by ERSP features obtained from varying frequency bands and time domains. The classification process was fulfilled using custom-written scripts via the FieldTrip Toolbox (version no: 20220104) integrated with the MVPA-light toolbox running under Matlab R2018b. The best performance was obtained by linear SVM which was fed by theta response (4-8 HZ) in the early time window (0.1-0.5 s) with 90% accuracy in the case of standard stimuli distinguished from novel stimuli. Delta responses also exhibit distinctive characteristics for standard and novel stimuli by running LDA (87% accuracy) and polynomial SVM (86% accuracy). These findings show that the delta and theta responses have contributed to detecting standard and novel sounds with remarkable performances of SVM and LDA.
P3.4 Event related delta oscillatory responses are decreased in patients with Alzheimer’s disease... more P3.4 Event related delta oscillatory responses are decreased in patients with Alzheimer’s disease B. Guntekin1, G.G. Yener2, D. Necioglu3, E. Tülay1, E. Basar1 1Istanbul Kultur University Brain Dynamics Cognition and Complex Systems Research Center, Istanbul, Turkey, 2Dokuz Eylul University Brain Dynamics and Multidisciplinary Research Center, Departments of Neurology and Neurosciences, Izmir, Turkey, 3Sisli Etfal State Hospital, Department of Neurology, Istanbul, Turkey
Machine Learning: Science and Technology, Jan 17, 2024
Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years ... more Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years thanks to their potential to enhance rehabilitation and control of prosthetic devices for individuals with motor disabilities. However, accurate classification of motor imagery signals remains a challenging task due to the high inter-subject variability and non-stationarity in the electroencephalogram (EEG) data. In the context of MI-BCIs, with limited data availability, the acquisition of EEG data can be difficult. In this study, several data augmentation techniques have been compared with the proposed data augmentation technique adaptive cross-subject segment replacement (ACSSR). This technique, in conjunction with the proposed deep learning framework, allows for a combination of similar subject pairs to take advantage of one another and boost the classification performance of MI-BCIs. The proposed framework features a multi-domain feature extractor based on common spatial patterns with a sliding window and a parallel two-branch convolutional neural network. The performance of the proposed methodology has been evaluated on the multi-class BCI Competition IV Dataset 2a through repeated 10-fold cross-validation. Experimental results indicated that the implementation of the ACSSR method (80.47%) in the proposed framework has led to a considerable improvement in the classification performance compared to the classification without data augmentation (77.63%), and other fundamental data augmentation techniques used in the literature. The study contributes to the advancements for the development of effective MI-BCIs by showcasing the ability of the ACSSR method to address the challenges in motor imagery signal classification tasks.
Attention-deficit/hyperactivity disorder (ADHD) is the most common neurodevelopmental disorder an... more Attention-deficit/hyperactivity disorder (ADHD) is the most common neurodevelopmental disorder and is characterized by symptoms of inattention and/or hyperactivity and impulsivity. In the current study, we obtained quantitative EEG (QEEG) recordings of 51 children aged between 6 and 12 years before the initiation of methylphenidate treatment. The relationship between changes in the scores of ADHD symptoms and initial QEEG features (power/power ratios values) were assessed. In addition, the children were classified as responder and nonresponder according to the ratio of their response to the medication (>25% improvement after medication). Logistic regression analyses were performed to analyze the accuracy of QEEG features for predicting responders. The findings indicate that patients with increased delta power at F8, theta power at Fz, F4, C3, Cz, T5, and gamma power at T6 and decreased beta powers at F8 and P3 showed more improvement in ADHD hyperactivity symptoms. In addition, i...
Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years ... more Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years thanks to their potential to enhance rehabilitation and control of prosthetic devices for individuals with motor disabilities. However, accurate classification of motor imagery signals remains a challenging task due to the high inter-subject variability and non-stationarity in the electroencephalogram (EEG) data. In the context of MI-BCIs, with limited data availability, the acquisition of EEG data can be difficult. In this study, several data augmentation techniques have been compared with the proposed data augmentation technique adaptive cross-subject segment replacement (ACSSR). This technique, in conjunction with the proposed deep learning framework, allows for a combination of similar subject pairs to take advantage of one another and boost the classification performance of MI-BCIs. The proposed framework features a multi-domain feature extractor based on common spatial patterns with a sliding window and a parallel two-branch convolutional neural network. The performance of the proposed methodology has been evaluated on the multi-class BCI Competition IV Dataset 2a through repeated 10-fold cross-validation. Experimental results indicated that the implementation of the ACSSR method (80.47%) in the proposed framework has led to a considerable improvement in the classification performance compared to the classification without data augmentation (77.63%), and other fundamental data augmentation techniques used in the literature. The study contributes to the advancements for the development of effective MI-BCIs by showcasing the ability of the ACSSR method to address the challenges in motor imagery signal classification tasks.
Objective. In the last decades, machine learning approaches have been widely used to distinguish ... more Objective. In the last decades, machine learning approaches have been widely used to distinguish Parkinson’s disease (PD) and many other neuropsychiatric diseases. They also speed up the clinicians and facilitate decision-making for several conditions with similar clinical symptoms. The current study attempts to detect PD with dementia (PDD) by event-related oscillations (EROs) during cognitive processing in two modalities, i.e. auditory and visual. Approach. The study was conducted to discriminate PDD from healthy controls (HC) using event-related phase-locking factors in slow frequency ranges (delta and theta) during visual and auditory cognitive tasks. Seventeen PDD and nineteen HC were included in the study, and linear discriminant analysis was used as a classifier. During classification analysis, multiple settings were implemented by using different sets of channels (overall, fronto-central and temporo-parieto-occipital (TPO) region), frequency bands (delta-theta combined, delta, theta, and low theta), and time of interests (0.1–0.7 s, 0.1–0.5 s and 0.1–0.3 s for delta, delta-theta combined; 0.1–0.4 s for theta and low theta) for spatial-spectral-temporal searchlight procedure. Main results. The classification performance results of the current study revealed that if visual stimuli are applied to PDD, the delta and theta phase-locking factor over fronto-central region have a remarkable contribution to detecting the disease, whereas if auditory stimuli are applied, the phase-locking factor in low theta over TPO and in a wider range of frequency (1–7 Hz) over the fronto-central region classify HC and PDD with better performances. Significance. These findings show that the delta and theta phase-locking factor of EROs during visual and auditory stimuli has valuable contributions to detecting PDD.
Lack of insight is a neurocognitive problem commonly encountered in patients with psychotic disor... more Lack of insight is a neurocognitive problem commonly encountered in patients with psychotic disorders that negatively affects treatment compliance and prognosis. Measurement of insight is based on self-report scales, which are limited due to subjectivity. This study aimed to determine the correlation between resting state beta and gamma power in 23 patients with schizophrenia and insight. It was observed that as beta and gamma power measured via qualitative electroencephalography (qEEG) increased the level of insight decreased. Negative correlation was found in F3, C3, Cz for gamma activity and in F3 and C3 for beta activity. This finding indicates that resting state qEEG could be used to evaluate the level of insight in patients with schizophrenia.
Unexpected events in the environment elicit the orienting response that protects humans from dang... more Unexpected events in the environment elicit the orienting response that protects humans from dangerous situations and there is great importance in identifying these events, especially in aging. The aims of the current study are attempting to find which classification model exhibits the best performance by means of event-related spectral perturbation (ERSP) features based on EEG and to understand which frequency bands, and time windows, contribute most to the classification of external stimuli. The data of 20 healthy elderly participants were included in the study and the 3-Stimulation auditory oddball paradigm was applied to participants. Different classifiers including Support Vector Machine (SVM) with Linear and Polynomial kernels, Linear Discriminant Analysis (LDA), and Naive Bayes were fed by ERSP features obtained from varying frequency bands and time domains. The classification process was fulfilled using custom-written scripts via the FieldTrip Toolbox (version no: 20220104) integrated with the MVPA-light toolbox running under Matlab R2018b. The best performance was obtained by linear SVM which was fed by theta response (4-8 HZ) in the early time window (0.1-0.5 s) with 90% accuracy in the case of standard stimuli distinguished from novel stimuli. Delta responses also exhibit distinctive characteristics for standard and novel stimuli by running LDA (87% accuracy) and polynomial SVM (86% accuracy). These findings show that the delta and theta responses have contributed to detecting standard and novel sounds with remarkable performances of SVM and LDA.
P3.4 Event related delta oscillatory responses are decreased in patients with Alzheimer’s disease... more P3.4 Event related delta oscillatory responses are decreased in patients with Alzheimer’s disease B. Guntekin1, G.G. Yener2, D. Necioglu3, E. Tülay1, E. Basar1 1Istanbul Kultur University Brain Dynamics Cognition and Complex Systems Research Center, Istanbul, Turkey, 2Dokuz Eylul University Brain Dynamics and Multidisciplinary Research Center, Departments of Neurology and Neurosciences, Izmir, Turkey, 3Sisli Etfal State Hospital, Department of Neurology, Istanbul, Turkey
Uploads
Papers by Elif Tülay