Brain Machine Interface

UIU Institutional Repository

    • Login
    View Item 
    •   UIU DSpace Home
    • School of Science and Engineering (SoSE)
    • Department of Computer Science and Engineering (CSE)
    • B.Sc Thesis/Project
    • View Item
    •   UIU DSpace Home
    • School of Science and Engineering (SoSE)
    • Department of Computer Science and Engineering (CSE)
    • B.Sc Thesis/Project
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Brain Machine Interface

    Thumbnail
    View/Open
    Thesis Book (3.130Mb)
    Date
    2018-03-16
    Author
    Miah, Md. Ochiuddin
    Rahman, Md. Mahfuzur
    Meno, Md. Rashed Khan
    Metadata
    Show full item record
    Abstract
    In last few decades fields like- Intelligence Systems for biosignals processing and modeling has developed extensively and this advancement has opened up many new windows of opportunities. Brain Machine Interface (BMI) is one of such opportunities. It is a technology which connects brain and machines directly in order to command and control the machines. This technology is about to revolutionize the health and rehabilitation industry. BMI system acquires signals generated from the brain as input. It then processes this signals to understand the user thought, intension, consequently which is used to generate command to operate machine. In this thesis, we explored the BMI technology and aimed to develop a system that can able to distinguish different human thoughts and use it as actions or commands for playing computer games. At the first part of this thesis, we obtained brain signals and extracted features from this signals. This features were analyzed to find out the informative patterns. We collected training and test data for designing two class classifier as well as for three class classifier. Two-class classifier classifies from right-hand movement and steady states. Whereas, three-class classifier classifies from right-hand movement, left-hand movement and steady states. At the second part of the thesis, we have worked with different classification methods including - OneR, Naïve Bayesian, C 4.5, and CART. For two-class classifier CART gave best performance, 91.3353% accuracy. But, C 4.5 did give very close performance to it, 90.8917% accuracy. For three-class classifier C 4.5 gave best performance, 65.6612% accuracy. At the last part of the thesis, we have considered overall performance and used C4.5 model to make the targeted application system. We made a virtual-ball movement controlling game, in which we can control the direction of movement of a virtual-ball using brain signals of voluntary movements. We used Emotiv SDK and Java language to develop a program to record the brain signals. We have used R-programming and weka tool for data visualization and model construction. We made an application system that can be used for rehabilitation as well as improvement of user well-being. User can exercise his mind to recover from attention deficiency. User can also increase his attention span via playing the game. It can be also used for gaming and entertainment purposes.
    URI
    http://dspace.uiu.ac.bd/handle/52243/201
    Collections
    • B.Sc Thesis/Project [82]

    Copyright 2003-2017 United International University
    Contact Us | Send Feedback
    Developed by UIU CITS
     

     

    Browse

    All of DSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

    My Account

    LoginRegister

    Copyright 2003-2017 United International University
    Contact Us | Send Feedback
    Developed by UIU CITS