Artificial neural network model for the prediction of obsessive-compulsive disorder treatment response

Giuliana Salomoni, Massimiliano Grassi, Paola Mosini, Patrizia Riva, Paolo Cavedini, Laura Bellodi

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Several patients with obsessive-compulsive disorder (OCD) who are refractory to adequate treatment with first-line treatments are considered treatment-resistant. Further surveys were to be implemented to explore the outcome predictors of the antiobsessional response. Such study was aimed at building a model suitable to predict the final outcome of a mixed OCD pharmacologic and nonpharmacologic treatment approaches. We studied 130 subjects with OCD who underwent pharmacologic (with selective serotonin reuptake inhibitors alone or with selective serotonin reuptake inhibitors and risperidone at low dosage) and/or behavioral therapy (using exposure and response prevention techniques). The following variables were used as predictors: symptoms dimension, as resulting from the Yale-Brown Obsessive-Compulsive Scale items factor analysis; neuropsychologic performances; and epidemiologic variables. The treatment response arising from 3 to 6 months of therapy was used as dependent variable. A conventional logistic regression was used to define a previsional model of treatment response and multilayer perceptrons and to supervise an artificial neural network technique. The 46.9% of the sample resulted to be refractory to treatment. Results obtained with the logistic regression model showed that the only predictors of treatment outcome are hoarding symptoms, repeating rituals, and counting compulsions. Furthermore, using all the variables considered in the models, multilayer perceptrons showed highly better predictive performance as compared with the logistic regression models (93.3% vs 61.5%, respectively, of correct classification of cases). Complex interactions between different clinical and neuropsychologic variables are involved in defining OCD treatment response profile, and nonlinear and interactive modeling strategies, that is, supervised artificial neural networks, seem to be more suitable to investigate this complexity than linear techniques.

Original languageEnglish
Pages (from-to)343-349
Number of pages7
JournalJournal of Clinical Psychopharmacology
Volume29
Issue number4
DOIs
Publication statusPublished - Aug 2009

Fingerprint

Neural Networks (Computer)
Obsessive-Compulsive Disorder
Logistic Models
Serotonin Uptake Inhibitors
Therapeutics
Implosive Therapy
Ceremonial Behavior
Risperidone
Statistical Factor Analysis

Keywords

  • Complexity
  • Logistic regression
  • Neural networks
  • Obsessive-compulsive disorder
  • Previsional models
  • Treatment predictors

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Pharmacology (medical)

Cite this

Artificial neural network model for the prediction of obsessive-compulsive disorder treatment response. / Salomoni, Giuliana; Grassi, Massimiliano; Mosini, Paola; Riva, Patrizia; Cavedini, Paolo; Bellodi, Laura.

In: Journal of Clinical Psychopharmacology, Vol. 29, No. 4, 08.2009, p. 343-349.

Research output: Contribution to journalArticle

Salomoni, Giuliana ; Grassi, Massimiliano ; Mosini, Paola ; Riva, Patrizia ; Cavedini, Paolo ; Bellodi, Laura. / Artificial neural network model for the prediction of obsessive-compulsive disorder treatment response. In: Journal of Clinical Psychopharmacology. 2009 ; Vol. 29, No. 4. pp. 343-349.
@article{68325ba23f9e4e358b779095be811967,
title = "Artificial neural network model for the prediction of obsessive-compulsive disorder treatment response",
abstract = "Several patients with obsessive-compulsive disorder (OCD) who are refractory to adequate treatment with first-line treatments are considered treatment-resistant. Further surveys were to be implemented to explore the outcome predictors of the antiobsessional response. Such study was aimed at building a model suitable to predict the final outcome of a mixed OCD pharmacologic and nonpharmacologic treatment approaches. We studied 130 subjects with OCD who underwent pharmacologic (with selective serotonin reuptake inhibitors alone or with selective serotonin reuptake inhibitors and risperidone at low dosage) and/or behavioral therapy (using exposure and response prevention techniques). The following variables were used as predictors: symptoms dimension, as resulting from the Yale-Brown Obsessive-Compulsive Scale items factor analysis; neuropsychologic performances; and epidemiologic variables. The treatment response arising from 3 to 6 months of therapy was used as dependent variable. A conventional logistic regression was used to define a previsional model of treatment response and multilayer perceptrons and to supervise an artificial neural network technique. The 46.9{\%} of the sample resulted to be refractory to treatment. Results obtained with the logistic regression model showed that the only predictors of treatment outcome are hoarding symptoms, repeating rituals, and counting compulsions. Furthermore, using all the variables considered in the models, multilayer perceptrons showed highly better predictive performance as compared with the logistic regression models (93.3{\%} vs 61.5{\%}, respectively, of correct classification of cases). Complex interactions between different clinical and neuropsychologic variables are involved in defining OCD treatment response profile, and nonlinear and interactive modeling strategies, that is, supervised artificial neural networks, seem to be more suitable to investigate this complexity than linear techniques.",
keywords = "Complexity, Logistic regression, Neural networks, Obsessive-compulsive disorder, Previsional models, Treatment predictors",
author = "Giuliana Salomoni and Massimiliano Grassi and Paola Mosini and Patrizia Riva and Paolo Cavedini and Laura Bellodi",
year = "2009",
month = "8",
doi = "10.1097/JCP.0b013e3181aba68f",
language = "English",
volume = "29",
pages = "343--349",
journal = "Journal of Clinical Psychopharmacology",
issn = "0271-0749",
publisher = "Lippincott Williams and Wilkins",
number = "4",

}

TY - JOUR

T1 - Artificial neural network model for the prediction of obsessive-compulsive disorder treatment response

AU - Salomoni, Giuliana

AU - Grassi, Massimiliano

AU - Mosini, Paola

AU - Riva, Patrizia

AU - Cavedini, Paolo

AU - Bellodi, Laura

PY - 2009/8

Y1 - 2009/8

N2 - Several patients with obsessive-compulsive disorder (OCD) who are refractory to adequate treatment with first-line treatments are considered treatment-resistant. Further surveys were to be implemented to explore the outcome predictors of the antiobsessional response. Such study was aimed at building a model suitable to predict the final outcome of a mixed OCD pharmacologic and nonpharmacologic treatment approaches. We studied 130 subjects with OCD who underwent pharmacologic (with selective serotonin reuptake inhibitors alone or with selective serotonin reuptake inhibitors and risperidone at low dosage) and/or behavioral therapy (using exposure and response prevention techniques). The following variables were used as predictors: symptoms dimension, as resulting from the Yale-Brown Obsessive-Compulsive Scale items factor analysis; neuropsychologic performances; and epidemiologic variables. The treatment response arising from 3 to 6 months of therapy was used as dependent variable. A conventional logistic regression was used to define a previsional model of treatment response and multilayer perceptrons and to supervise an artificial neural network technique. The 46.9% of the sample resulted to be refractory to treatment. Results obtained with the logistic regression model showed that the only predictors of treatment outcome are hoarding symptoms, repeating rituals, and counting compulsions. Furthermore, using all the variables considered in the models, multilayer perceptrons showed highly better predictive performance as compared with the logistic regression models (93.3% vs 61.5%, respectively, of correct classification of cases). Complex interactions between different clinical and neuropsychologic variables are involved in defining OCD treatment response profile, and nonlinear and interactive modeling strategies, that is, supervised artificial neural networks, seem to be more suitable to investigate this complexity than linear techniques.

AB - Several patients with obsessive-compulsive disorder (OCD) who are refractory to adequate treatment with first-line treatments are considered treatment-resistant. Further surveys were to be implemented to explore the outcome predictors of the antiobsessional response. Such study was aimed at building a model suitable to predict the final outcome of a mixed OCD pharmacologic and nonpharmacologic treatment approaches. We studied 130 subjects with OCD who underwent pharmacologic (with selective serotonin reuptake inhibitors alone or with selective serotonin reuptake inhibitors and risperidone at low dosage) and/or behavioral therapy (using exposure and response prevention techniques). The following variables were used as predictors: symptoms dimension, as resulting from the Yale-Brown Obsessive-Compulsive Scale items factor analysis; neuropsychologic performances; and epidemiologic variables. The treatment response arising from 3 to 6 months of therapy was used as dependent variable. A conventional logistic regression was used to define a previsional model of treatment response and multilayer perceptrons and to supervise an artificial neural network technique. The 46.9% of the sample resulted to be refractory to treatment. Results obtained with the logistic regression model showed that the only predictors of treatment outcome are hoarding symptoms, repeating rituals, and counting compulsions. Furthermore, using all the variables considered in the models, multilayer perceptrons showed highly better predictive performance as compared with the logistic regression models (93.3% vs 61.5%, respectively, of correct classification of cases). Complex interactions between different clinical and neuropsychologic variables are involved in defining OCD treatment response profile, and nonlinear and interactive modeling strategies, that is, supervised artificial neural networks, seem to be more suitable to investigate this complexity than linear techniques.

KW - Complexity

KW - Logistic regression

KW - Neural networks

KW - Obsessive-compulsive disorder

KW - Previsional models

KW - Treatment predictors

UR - http://www.scopus.com/inward/record.url?scp=68449101172&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=68449101172&partnerID=8YFLogxK

U2 - 10.1097/JCP.0b013e3181aba68f

DO - 10.1097/JCP.0b013e3181aba68f

M3 - Article

C2 - 19593173

AN - SCOPUS:68449101172

VL - 29

SP - 343

EP - 349

JO - Journal of Clinical Psychopharmacology

JF - Journal of Clinical Psychopharmacology

SN - 0271-0749

IS - 4

ER -