1 AIT Asian Institute of Technology

Investigating the effectiveness of transfer learning in a motor imagery game

AuthorPongkorn Settasompop
Call NumberAIT Thesis no.CS-23-02
Subject(s)Electroencephalography
Brain-computer interfaces
Transfer learning (Machine learning)
NoteA thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science
PublisherAsian Institute of Technology
AbstractThis paper investigates the effectiveness of transfer learning in the BCI motor imagery system. In this study, an online EEG-based MI-BCI system with a Unity game as feedback was developed to assess transfer learning. If the participant successfully completes a task, the car is moved. Seven participants participated in the execution and imagery EEG signal recordings. Seven EEG recordings were used to train the 1D-CNN. Seven individuals were requested to participate in an online experiment that required the collection of additional EEG data for transfer learning. Our classifiers are more efficient in both execution and im age classification. In addition, effective classification was supplied for the online system. We conclude that MI-BCI-Unity is the next possible upgrade to the MI-BCI system.
Year2023
TypeThesis
SchoolSchool of Engineering and Technology
DepartmentDepartment of Information and Communications Technologies (DICT)
Academic Program/FoSComputer Science (CS)
Chairperson(s)Chaklam Silpasuwanchai
Examination Committee(s)Dailey, Matthew N.;Mongkol Ekpanyapong
Scholarship Donor(s)Royal Thai Government
DegreeThesis (M. Sc.) - Asian Institute of Technology, 2023


Usage Metrics
View Detail0
Read PDF0
Download PDF0