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

Evaluating clustering methods for psychiatric diagnosis.

J E Mezzich

    Biological Psychiatry
    |April 1, 1978
    PubMed
    Summary
    This summary is machine-generated.

    This study evaluated clustering methods for psychiatric diagnoses. Nearest centroid and hierarchical methods performed best, outperforming mixture analysis for grouping patients based on psychopathological variables.

    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

    Promoting healthy lives and well-being for all: The contribution of the International College of Person-Centered Medicine (ICPCM).

    Psychiatrike = Psychiatriki·2018
    Same author

    A comprehensive and individualized psychiatric interviewing training program.

    Academic psychiatry : the journal of the American Association of Directors of Psychiatric Residency Training and the Association for Academic Psychiatry·2014
    Same author

    Person-centred medicine and mental health.

    Epidemiology and psychiatric sciences·2012
    Same author

    [A validation study of the Personal Health Scale in Argentina and Venezuela].

    Actas espanolas de psiquiatria·2010
    Same author

    A bibliometric study of the use of the classification and diagnostic systems in psychiatry over the last 25 years.

    Psychopathology·2008
    Same author

    The multicultural quality of life index in Argentina: a validation study.

    Actas espanolas de psiquiatria·2007
    Same journal

    Impact of Transcranial Direct Current Stimulation-Induced Electric Fields on Slowing Cognitive Decline in Older Adults With Mild Cognitive Impairment or Remitted Major Depressive Disorder: An Analysis of the PACt-MD Randomized Clinical Trial.

    Biological psychiatry·2026
    Same journal

    Remembering Jon-Kar Zubieta, M.D., Ph.D.

    Biological psychiatry·2026
    Same journal

    Kappa opioid receptor availability in borderline personality disorder: An in-vivo investigation with [<sup>11</sup>C]EKAP PET imaging.

    Biological psychiatry·2026
    Same journal

    From Satiety to Substance Use: Neural Mechanisms of GLP-1 Signaling in Appetite and Reward.

    Biological psychiatry·2026
    Same journal

    Distinct and Shared Molecular Mechanisms Underlie Morphological-Functional Overcoupling and Undercoupling in Major Depressive Disorder.

    Biological psychiatry·2026
    Same journal

    Dynamic Brain States With Cannabis Intoxication: Beyond "More Is Better" in Interpreting Brain Connectivity.

    Biological psychiatry·2026
    See all related articles

    Area of Science:

    • Psychiatry
    • Computer Science
    • Data Science

    Background:

    • Accurate psychiatric diagnosis relies on classifying patients into distinct categories.
    • Clustering algorithms offer computational approaches to identify patient groupings based on shared characteristics.

    Purpose of the Study:

    • To empirically evaluate major clustering approaches using psychiatric diagnostic data.
    • To compare the performance of ten computerized clustering methods on a dataset of archetypal psychiatric patients.

    Main Methods:

    • Eight-eight archetypal patients representing four diagnostic categories were created using 17 psychopathological variables.
    • Ten computerized clustering methods were applied to the patient data to form new groupings.
    • Evaluative criteria included concordance with original data structure and clustering replicability.

    Related Experiment Videos

    Main Results:

    • Significant differences were observed in the performance of various clustering methods.
    • Nearest centroid sorting, complete linkage, and centroid linkage hierarchical methods ranked highest.
    • Multivariate normal mixture analysis and facial representation of multidimensional points yielded the poorest results.

    Conclusions:

    • The choice of clustering method significantly impacts the accuracy of psychiatric patient classification.
    • Nearest centroid and hierarchical clustering methods demonstrate superior performance for psychiatric diagnostic data.
    • These findings suggest specific algorithms are more suitable for analyzing complex biological and psychosocial datasets in psychiatry.