Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Direct speech feature estimation using an iterative EM algorithm for vocal fold pathology detection

L Gavidia-Ceballos1, J H Hansen

  • 1Department of Biomedical Engineering, Duke University, Durham, NC 37708-0291, USA. lgavadia@usb.ve

IEEE Transactions on Bio-Medical Engineering
|April 1, 1996
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Frontoparietal network topology as a neural marker of musical perceptual abilities.

Nature communications·2024
Same author

Ventricular function and ventriculo-arterial coupling after palliation of hypoplastic left heart syndrome: A comparative study with Fontan patients with LV morphology.

International journal of cardiology·2016
Same author

Effects of organic plant oils and role of oxidation on nutrient utilization in juvenile rainbow trout (Oncorhynchus mykiss).

Animal : an international journal of animal bioscience·2012
Same author

MicroRNA identity and abundance in porcine skeletal muscles determined by deep sequencing.

Animal genetics·2009
Same author

A nonlinear operator-based speech feature analysis method with application to vocal fold pathology assessment.

IEEE transactions on bio-medical engineering·1998
Same author

Temperature distribution in expiratory speaking flow, and early detection of vocal fold pathology.

Journal of medical engineering & technology·1997
Same journal

Assessment of skin stiffness in systemic sclerosis using optical coherence elastography: A comparative study with histology and clinical parameters.

IEEE transactions on bio-medical engineering·2026
Same journal

Modeling Dyadic Interdependence in Endocrine Functioning: A Multilevel Machine Learning Study of Adults with Cancer and Their Caregivers.

IEEE transactions on bio-medical engineering·2026
Same journal

A Kalman Filter-Based Framework for Granger Causality Assessment: Application in Tracking Maternal-Fetal Heart Rate Coupling.

IEEE transactions on bio-medical engineering·2026
Same journal

Enhancing Volumetric Imaging in Linear-Array Photoacoustic Tomography: multiview fusion with deep learning.

IEEE transactions on bio-medical engineering·2026
Same journal

Robust Rule-based Heuristic Assistance Strategy for a Semi-Active Shoulder Exoskeleton Used in Overhead Work.

IEEE transactions on bio-medical engineering·2026
Same journal

Highly Accelerated 1-mm Isotropic 3D Chemical Exchange Saturation Transfer MRI Using Wave-Co-CAIPI at 5 Tesla.

IEEE transactions on bio-medical engineering·2026
See all related articles

A new speech analysis algorithm accurately detects vocal fold pathology using enhanced spectral features (ESPC). This method, based on the estimation-maximization (EM) algorithm, offers a noninvasive approach for diagnosing conditions like vocal fold cancer.

Area of Science:

  • Speech processing
  • Biomedical engineering
  • Acoustic analysis

Background:

  • Vocal fold pathology detection is crucial for diagnosis.
  • Existing methods may struggle with incomplete glottal closure.
  • Need for quantitative, noninvasive analysis techniques.

Purpose of the Study:

  • To develop a speech parameter estimation algorithm for vocal fold pathology analysis.
  • To formulate a stochastic model for characterizing healthy and pathological speech.
  • To introduce and validate novel spectral features for pathology detection.

Main Methods:

  • Iterative maximum-likelihood (ML) estimation using the estimation-maximization (EM) algorithm.
  • Estimation of enhanced-spectral-pathology component (ESPC) feature.

Related Experiment Videos

  • Development of a hidden Markov model (HMM) classifier using ESPC, MAPV, and WSLOPE features.
  • Utilized log Mel-frequency filter bank coefficients for ESPC parameterization.

    • Main Results:

    • The enhanced-spectral-pathology component (ESPC) feature consistently differentiates between healthy and pathological conditions.
    • Mean-area-peak-value (MAPV) and weighted-slope (WSLOPE) indexes derived from ESPC are meaningful indicators of pathology.
    • A hidden Markov model (HMM) classifier achieved high detection rates (88.7% pathology, 92.8% healthy) using fine spectral representation of ESPC.
    • The method bypasses the need for direct glottal flow waveform estimation, overcoming limitations of incomplete glottal closure.

    Conclusions:

    • The proposed algorithm provides a quantitative, noninvasive method for vocal fold pathology analysis and detection.
    • Enhanced-spectral-pathology component (ESPC) analysis offers a robust approach for characterizing speech production abnormalities.
    • The method demonstrates significant potential for clinical application in diagnosing vocal fold conditions.