• About
    • Login
    View Item 
    •   Institutional Repository Home
    • Electronic Theses and Dissertations
    • Electronic Theses and Dissertations
    • View Item
    •   Institutional Repository Home
    • Electronic Theses and Dissertations
    • Electronic Theses and Dissertations
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Browse

    All of Institutional RepositoryCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsDepartmentThis CollectionBy Issue DateAuthorsTitlesSubjectsDepartment

    My Account

    LoginRegister

    Analysis of speech features as potential indicators for depression and high risk suicide and possible predictors for the Hamilton Depression rating (HAMD) and Beck Depression Inventory scale (BDI-II)

    Nik Hashim, Nik Nur Wahidah
    : https://etd.library.vanderbilt.edu/etd-03192014-065659
    http://hdl.handle.net/1803/10891
    : 2014-03-29

    Abstract

    Patients who are diagnosed with depression without appropriate clinical recognition of their hidden suicidal tendencies are at elevated risk of making suicide attempts. An important clinical problem remains the differentiation between non-suicidal and more lethal episodes of depression. It requires clinicians to use specific interviewing approaches, sometimes relying on their intuition, and to deploy specialized skills developed through formal education, clinical training and clinical experience. Conventional methods are generally performed according to a series of questionnaires and rating that requires non-static and time-consuming information gathering process. In an effort to find a reliable method that could assist clinicians in risk assessment, information in the speech signal has been found to contain characteristic changes associated with psychological states. This dissertation shows the ability of speech timing based measures to discriminate the high risk suicidal speech from the depressed speech using only a single and/or two combinations of features to produce an effective classifier performance. The speech timing based measures were also shown to be robust across different data sets despite the less than ideal recording conditions and different equipment used for each database. The second analysis demonstrates the effectiveness of using acoustic measurements as a possible means to predict ratings from well-known medical diagnostic tools known as the Hamilton Depression Scale (HAMD) and Beck Depression Inventory (BDI-II) with an error less than 10% of the actual score. For the third analysis, investigation of the characteristics of root mean square amplitude modulation (RMS AM) in the speech of depressed and near-term suicidal patients identified that a feature combination of range and skewness yielded a significant discriminator for male speech. The final study demonstrates the use of Power Spectral Density (PSD) feature in measuring the significance between high risk suicidal and depressed patients, and the significant improvement in patients’ psychological state as they progress from one recording session to the next, collected a few days after receiving treatment.
    Show full item record

    Files in this item

    Icon
    Name:
    NikNur_Thesis.pdf
    Size:
    3.461Mb
    Format:
    PDF
    View/Open

    This item appears in the following collection(s):

    • Electronic Theses and Dissertations

    Connect with Vanderbilt Libraries

    Your Vanderbilt

    • Alumni
    • Current Students
    • Faculty & Staff
    • International Students
    • Media
    • Parents & Family
    • Prospective Students
    • Researchers
    • Sports Fans
    • Visitors & Neighbors

    Support the Jean and Alexander Heard Libraries

    Support the Library...Give Now

    Gifts to the Libraries support the learning and research needs of the entire Vanderbilt community. Learn more about giving to the Libraries.

    Become a Friend of the Libraries

    Quick Links

    • Hours
    • About
    • Employment
    • Staff Directory
    • Accessibility Services
    • Contact
    • Vanderbilt Home
    • Privacy Policy