Document Details

Document Type : Thesis 
Document Title :
Classification and Assessment of Cognitive Skills from EEG and NIRS Signals
تصنيف وتقييم المهارات المعرفية من إشارات تخطيط كهربية الدماغ ومطيافية الأشعة تحت الحمراء القريبة
 
Subject : Faculty of Engineering 
Document Language : Arabic 
Abstract : Human mind can be directly connected with the computers through a new technology known as Brain Computer Interface (BCI). Electroencephalography (EEG) and Near Infrared Spectroscopy (NIRS) based BCI enables to interact the people with the surrounding world through brain signals noninvasively. This method of reading the mind through physiological signals by EEG and NIRS sensors has made significant progress in neurological science and motor control research. The BCI system can record, analyze, and translate the system input acquired from the brain in terms of commands. These commands can further be used to actuate external devices of choice according to the user’s mind. The BCI is emerging as one of the powerful tools in realistic biomedical applications such as rehabilitation, cognitive processes, prosthetics, and many neuro-feedback functional activities. However, the functionality of BCI relies upon the recognition and classification of brain signals for discriminating task and resting activities of the brain. In this thesis, we have developed two algorithms for assessment and classification of EEG and NIRS alone and combined as hybrid (EEG+NIRS) signals recognizing brain activities under the given tasks. We have tested our classifiers from open source EEG-NIRS dataset. The dataset is consisting of EEG and NIRS simultaneously recordings acquired from 26 healthy participants during word generation (WG) tasks. By implementing our algorithms which are based on SVM (Support Vector Machine) and Linear Discriminant Analysis (LDA) methods, the signals of rest and task events were classified more precisely with SVM than the LDA algorithm. We have achieved an average classification accuracy peak of 85 %, 84 % and 78 % Hybrid, EEG and NIRS respectively for SVM comparing to LDA 81 %, 77 % and 75 % Hybrid, EEG and NIRS respectively with the dataset. 
Supervisor : Dr. Prahlad Rao Kalyanrao 
Thesis Type : Master Thesis 
Publishing Year : 1442 AH
2021 AD 
Co-Supervisor : Dr. Muhammad Moinuddin 
Added Date : Wednesday, July 7, 2021 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
حسين سالم الحريبيAL-Huraibi, Hussein SalemResearcherMaster 

Files

File NameTypeDescription
 47085.pdf pdf 

Back To Researches Page