A neural network model for combining clinical predictors of antidepressant response in mood disorders

Alessandro Serretti, Raffaella Zanardi, Laura Mandelli, Enrico Smeraldi, Cristina Colombo

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Artificial neural networks (ANN) represent a promising tool for combining multiple predictors in complex diseases. Antidepressant response in mood disorders is a typical complex phenomenon were a number of predictors influence outcome under non-linear interactions. In the present study we tested a neural network strategy for antidepressant outcome in subjects affected by major depression. One hundred and forty-five never reported depressed inpatients were included in this study (major depressives/bipolars: 111/34). A multi layer perceptron network composed of 1 hidden layer with 13 nodes was chosen. The network was performed on the sample of 145 cases divided as follows: train 73 + verify 36 + test 36. Correlation of predicted versus observed response was 0.46 in the test (independent) sample that corresponds to 21% of variance explained. Number of episodes, side effects, delusional features, baseline HAM-D, length of current episode, lithium augmentation, current medical condition and personality disorders were the main factors identified by the model. Sex, age at onset, polarity, plasma level and baseline VAS score were part of the model but with a lower rank. Overall, our findings suggest that neural networks could be a valid technique for the analysis of psychopharmacology studies. The complex interactions modelled through ANN may be eventually applied at the clinical level for the individualized therapy.

Original languageEnglish
Pages (from-to)239-245
Number of pages7
JournalJournal of Affective Disorders
Volume98
Issue number3
DOIs
Publication statusPublished - Mar 2007

Fingerprint

Neural Networks (Computer)
Mood Disorders
Antidepressive Agents
Psychopharmacology
Personality Disorders
Lithium
Age of Onset
Inpatients
Depression
Therapeutics

Keywords

  • Bipolar disorder
  • Major depressive disorder
  • Neural network
  • Outcome predictors

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Behavioral Neuroscience
  • Biological Psychiatry
  • Neurology
  • Psychology(all)

Cite this

A neural network model for combining clinical predictors of antidepressant response in mood disorders. / Serretti, Alessandro; Zanardi, Raffaella; Mandelli, Laura; Smeraldi, Enrico; Colombo, Cristina.

In: Journal of Affective Disorders, Vol. 98, No. 3, 03.2007, p. 239-245.

Research output: Contribution to journalArticle

Serretti, Alessandro ; Zanardi, Raffaella ; Mandelli, Laura ; Smeraldi, Enrico ; Colombo, Cristina. / A neural network model for combining clinical predictors of antidepressant response in mood disorders. In: Journal of Affective Disorders. 2007 ; Vol. 98, No. 3. pp. 239-245.
@article{1834a6c703bc4f638a1c67648812d767,
title = "A neural network model for combining clinical predictors of antidepressant response in mood disorders",
abstract = "Artificial neural networks (ANN) represent a promising tool for combining multiple predictors in complex diseases. Antidepressant response in mood disorders is a typical complex phenomenon were a number of predictors influence outcome under non-linear interactions. In the present study we tested a neural network strategy for antidepressant outcome in subjects affected by major depression. One hundred and forty-five never reported depressed inpatients were included in this study (major depressives/bipolars: 111/34). A multi layer perceptron network composed of 1 hidden layer with 13 nodes was chosen. The network was performed on the sample of 145 cases divided as follows: train 73 + verify 36 + test 36. Correlation of predicted versus observed response was 0.46 in the test (independent) sample that corresponds to 21{\%} of variance explained. Number of episodes, side effects, delusional features, baseline HAM-D, length of current episode, lithium augmentation, current medical condition and personality disorders were the main factors identified by the model. Sex, age at onset, polarity, plasma level and baseline VAS score were part of the model but with a lower rank. Overall, our findings suggest that neural networks could be a valid technique for the analysis of psychopharmacology studies. The complex interactions modelled through ANN may be eventually applied at the clinical level for the individualized therapy.",
keywords = "Bipolar disorder, Major depressive disorder, Neural network, Outcome predictors",
author = "Alessandro Serretti and Raffaella Zanardi and Laura Mandelli and Enrico Smeraldi and Cristina Colombo",
year = "2007",
month = "3",
doi = "10.1016/j.jad.2006.08.008",
language = "English",
volume = "98",
pages = "239--245",
journal = "Journal of Affective Disorders",
issn = "0165-0327",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - A neural network model for combining clinical predictors of antidepressant response in mood disorders

AU - Serretti, Alessandro

AU - Zanardi, Raffaella

AU - Mandelli, Laura

AU - Smeraldi, Enrico

AU - Colombo, Cristina

PY - 2007/3

Y1 - 2007/3

N2 - Artificial neural networks (ANN) represent a promising tool for combining multiple predictors in complex diseases. Antidepressant response in mood disorders is a typical complex phenomenon were a number of predictors influence outcome under non-linear interactions. In the present study we tested a neural network strategy for antidepressant outcome in subjects affected by major depression. One hundred and forty-five never reported depressed inpatients were included in this study (major depressives/bipolars: 111/34). A multi layer perceptron network composed of 1 hidden layer with 13 nodes was chosen. The network was performed on the sample of 145 cases divided as follows: train 73 + verify 36 + test 36. Correlation of predicted versus observed response was 0.46 in the test (independent) sample that corresponds to 21% of variance explained. Number of episodes, side effects, delusional features, baseline HAM-D, length of current episode, lithium augmentation, current medical condition and personality disorders were the main factors identified by the model. Sex, age at onset, polarity, plasma level and baseline VAS score were part of the model but with a lower rank. Overall, our findings suggest that neural networks could be a valid technique for the analysis of psychopharmacology studies. The complex interactions modelled through ANN may be eventually applied at the clinical level for the individualized therapy.

AB - Artificial neural networks (ANN) represent a promising tool for combining multiple predictors in complex diseases. Antidepressant response in mood disorders is a typical complex phenomenon were a number of predictors influence outcome under non-linear interactions. In the present study we tested a neural network strategy for antidepressant outcome in subjects affected by major depression. One hundred and forty-five never reported depressed inpatients were included in this study (major depressives/bipolars: 111/34). A multi layer perceptron network composed of 1 hidden layer with 13 nodes was chosen. The network was performed on the sample of 145 cases divided as follows: train 73 + verify 36 + test 36. Correlation of predicted versus observed response was 0.46 in the test (independent) sample that corresponds to 21% of variance explained. Number of episodes, side effects, delusional features, baseline HAM-D, length of current episode, lithium augmentation, current medical condition and personality disorders were the main factors identified by the model. Sex, age at onset, polarity, plasma level and baseline VAS score were part of the model but with a lower rank. Overall, our findings suggest that neural networks could be a valid technique for the analysis of psychopharmacology studies. The complex interactions modelled through ANN may be eventually applied at the clinical level for the individualized therapy.

KW - Bipolar disorder

KW - Major depressive disorder

KW - Neural network

KW - Outcome predictors

UR - http://www.scopus.com/inward/record.url?scp=33846411843&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846411843&partnerID=8YFLogxK

U2 - 10.1016/j.jad.2006.08.008

DO - 10.1016/j.jad.2006.08.008

M3 - Article

VL - 98

SP - 239

EP - 245

JO - Journal of Affective Disorders

JF - Journal of Affective Disorders

SN - 0165-0327

IS - 3

ER -