Yazar "Canbaz Kabay, Sibel" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim
    Yayın
    EEG based environment classification during cognitive task of multiple sclerosis patients
    (Institute of Electrical and Electronics Engineers Inc., 2022) Şaşmaz Karacan, Seda; Saraoğlu, Hamdi Melih; Canbaz Kabay, Sibel; Akdağ, Gönül; Keskinkılıç, Cahit; Tosun, Mustafa
    Multiple sclerosis (MS) is a neurodegenerative central nervous system disease in which the tissues in the brain, cerebellum, brain stem, and spinal cord are damaged as a result of the immune system disorder. The aim of this study is to classify the environment from the EEG signals recorded during the cognitive task in the computer and virtual reality environment of MS patients and healthy volunteers. Multilayer perceptron (MLP), k-nearest neighbors algorithm (kNN), and Support Vector Machine (SVM) classifiers' performances are compared using EEG signals during a cognitive task of 11 MS patients and 28 healthy volunteers. EEG signals of volunteers are separated into alpha, beta, gamma, delta, and theta subbands with Wavelet Daubechies (db2). Spectral and statistical features of the subbands are extracted. The most important features are determined by the Recursive Feature Elimination (RFE) algorithm. Training and testing data are separated by Leave-One-Out Cross-Validation. While the best environment classification for healthy volunteers is 91.07% accuracy with the SVM classifier, the best classification performance for volunteers with MS is 95.45% accuracy with the kNN classifier.
  • Küçük Resim
    Yayın
    EEG-based mental workload estimation of multiple sclerosis patients
    (Springer London Ltd, 2023) Şaşmaz Karacan, Seda; Saraoğlu, Hamdi Melih; Canbaz Kabay, Sibel; Akdağ, Gönül; Keskinkılıç, Cahit; Tosun, Mustafa
    The amount of mental capacity required by individuals to complete any task is defined as mental workload. It is important to determine the appropriate level in order not to impose too much mental workload on individuals or not to create unnecessary human resources for the completion of a task. Multiple Sclerosis (MS) is an autoimmune neurodegenerative central nervous system disease that activates the acquired and innate immune systems due to the interaction between genetic and environmental factors and manifests itself with different neurological symptoms. This study aims to classify the mental workload level of MS patients as low, medium, or high from EEG signals during cognitive tasks in computer and virtual reality environments and to compare them with a healthy group performing the same tasks. In this study, the mental workload level of 45 volunteers is estimated by using EEG signals and NASA-Raw Task Load Index questionnaire results in 3 cognitive tasks in computer and virtual reality environments. The three-level mental workload classification accuracy in MS patients with the Support Vector Machine classifier is 96.08% and 94.12% for computer and virtual reality environments, respectively. For healthy volunteers, classification accuracy is 95.24% and 94.05% in computer and virtual reality environments, respectively. In the study, mental workload research was conducted for the first time from EEG signals of MS patients obtained during cognitive tasks in computer and virtual reality environments.
  • Küçük Resim
    Yayın
    Understanding the relationship between rosemary odor and mental workload through deep learning
    (Elsevier Ltd, 2025) Şahin Sadık, Evin; Saraoğlu, Hamdi Melih; Canbaz Kabay, Sibel; Keskinkılıç, Cahit
    This research explores the novel application of aromatic odors, specifically rosemary, in reducing mental workload, employing deep learning methods to analyze electroencephalogram (EEG) signals without feature extraction. Thirty volunteers participated in five neuropsychological tests while being exposed to the aroma of rosemary. The EEG signals recorded during the performance of these tasks were analyzed using deep learning methods to classify mental workload. Deep learning algorithms such as Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) were employed to classify mental workload directly from EEG signals. The analysis revealed that volunteers exposed to the rosemary odor showed decreased error rates and increased test success and learning scores, in comparison to a condition without odor. The classification of mental workload under rosemary odor exposure was achieved with a high accuracy rate of 97.11% in both deep learning algorithms. This study presents a novel approach by combining olfactory stimulation and EEG-based mental workload classification through deep learning. These findings suggest that rosemary odor may reduce mental workload and that raw EEG signals can be effectively analyzed using deep learning without manual feature engineering.