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

Neural network subtyping of depression

T M Florio1, G Parker, M P Austin

  • 1Psychiatry Unit, Prince of Wales Hospital, Randwick, New South Wales, Australia.

The Australian and New Zealand Journal of Psychiatry
|November 7, 1998
PubMed
Summary

Neural networks effectively classify depressive subtypes by analyzing psychomotor disturbance (PMD) and endogeneity symptoms. This non-linear approach supports melancholic depression as a distinct disorder, advancing psychiatric classification.

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

Changes in white matter microstructure and MRI-derived cerebral blood flow after 1-week of exercise training.

Scientific reports·2021
Same author

Refined diagnostic criteria for the bipolar disorders: phase two of the AREDOC project.

Acta psychiatrica Scandinavica·2021
Same author

Acoustic shock: an update review.

The Journal of laryngology and otology·2020
Same author

Paradigms lost, gained and retained.

Acta psychiatrica Scandinavica·2020
Same author

Differentiating clinical and non-clinical depression: a heuristic study offering a template for extension studies.

Acta psychiatrica Scandinavica·2019
Same author

How to choose an antidepressant medication.

Acta psychiatrica Scandinavica·2018

Area of Science:

  • Psychiatry
  • Computational Neuroscience
  • Machine Learning

Background:

  • The DSM-III-R definition of melancholia relies on specific symptom clusters.
  • Distinguishing between melancholic and non-melancholic depression is crucial for accurate diagnosis and treatment.
  • The independent contributions of psychomotor disturbance (PMD) and endogeneity symptoms to melancholia require further investigation.

Purpose of the Study:

  • To assess the utility of a neural network classification strategy.
  • To evaluate the distinct roles of PMD and endogeneity symptoms in defining melancholia.
  • To compare non-linear neural network models with traditional linear analysis.

Main Methods:

  • A dataset of 407 depressed patients was analyzed.
  • Included were 17 endogeneity symptoms and the 18-item CORE measure for behaviorally rated PMD.

Related Experiment Videos

  • Multilayer perception neural networks and linear discriminant function analysis were employed.

    • Main Results:

    • Models using only PMD or endogeneity symptoms showed comparable classification success.
    • Non-linear models integrating both PMD and endogeneity symptom scores yielded the highest classification accuracy.
    • The non-linear model outperformed the linear discriminant analysis.

    Conclusions:

    • Non-linear modeling with neural networks offers a promising approach for psychiatric diagnostic taxonomy.
    • Findings support a binary classification of depression, viewing melancholic and non-melancholic subtypes as separate disorders.
    • This methodology has potential applications in clinical decision-making for psychiatric disorders.