Can Machine Learning help us in dealing with treatment resistant depression? A review

Alessandro Pigoni, Giuseppe Delvecchio, Domenico Madonna, Cinzia Bressi, Jair Soares, Paolo Brambilla

Research output: Contribution to journalArticle

Abstract

Background: About one third of patients treated with antidepressant do not show sufficient symptoms relief and up to 15% of patients remain symptomatic even after multiple trials are applied, configuring a state called treatment resistant depression (TRD). A clear definition of this state and the understanding of underlying mechanisms contributing to chronic disability caused by major depressive disorder is still unknown. Therefore, Machine Learning (ML) techniques emerged in the last years as interesting approaches to deal with such complex problems. Methods: We performed a bibliographic search on Pubmed, Google Scholar and Medline of clinical, imaging, genetic and EEG ML classification studies on treatment-responding depression and TRD as well as studies trying to predict response to a specific treatment in already established TRD. The inclusion criteria were met by eleven studies. Seven focused on the definition of predictors of TRD onset while four attempted to predict the response to specific treatments in TRD. Results: The results showed that it seems possible to classify between responders MDD and TRD with good accuracies based on clinical variables. Moreover, some studies reported the possibility of using EEG measures to predict response to different pharmacological and non-pharmacological treatments in established TRD. Limitations: The definition of TRD, the selection of variables together with ML algorithms and pipelines varies across the studies, ultimately determining the unfeasibility to implement these models in clinical practice. Conclusions: The findings suggest that ML could be a valid approach to increase our understanding of TRD and to better classify and stratify this disorder, which may ultimately help clinicians in the assessment of major depressive disorders.

Original languageEnglish
Pages (from-to)21-26
Number of pages6
JournalJournal of Affective Disorders
Volume259
DOIs
Publication statusPublished - Dec 1 2019

Fingerprint

Treatment-Resistant Depressive Disorder
Major Depressive Disorder
Electroencephalography
Machine Learning
Therapeutics
PubMed
Antidepressive Agents
Pharmacology
Depression

Keywords

  • Depression
  • EEG
  • Imaging
  • Machine Learning
  • Major depressive disorder
  • Treatment resistant depression

ASJC Scopus subject areas

  • Clinical Psychology
  • Psychiatry and Mental health

Cite this

Can Machine Learning help us in dealing with treatment resistant depression? A review. / Pigoni, Alessandro; Delvecchio, Giuseppe; Madonna, Domenico; Bressi, Cinzia; Soares, Jair; Brambilla, Paolo.

In: Journal of Affective Disorders, Vol. 259, 01.12.2019, p. 21-26.

Research output: Contribution to journalArticle

Pigoni, Alessandro ; Delvecchio, Giuseppe ; Madonna, Domenico ; Bressi, Cinzia ; Soares, Jair ; Brambilla, Paolo. / Can Machine Learning help us in dealing with treatment resistant depression? A review. In: Journal of Affective Disorders. 2019 ; Vol. 259. pp. 21-26.
@article{09282c0dcaf048ab8a303461fec4f8bd,
title = "Can Machine Learning help us in dealing with treatment resistant depression? A review",
abstract = "Background: About one third of patients treated with antidepressant do not show sufficient symptoms relief and up to 15{\%} of patients remain symptomatic even after multiple trials are applied, configuring a state called treatment resistant depression (TRD). A clear definition of this state and the understanding of underlying mechanisms contributing to chronic disability caused by major depressive disorder is still unknown. Therefore, Machine Learning (ML) techniques emerged in the last years as interesting approaches to deal with such complex problems. Methods: We performed a bibliographic search on Pubmed, Google Scholar and Medline of clinical, imaging, genetic and EEG ML classification studies on treatment-responding depression and TRD as well as studies trying to predict response to a specific treatment in already established TRD. The inclusion criteria were met by eleven studies. Seven focused on the definition of predictors of TRD onset while four attempted to predict the response to specific treatments in TRD. Results: The results showed that it seems possible to classify between responders MDD and TRD with good accuracies based on clinical variables. Moreover, some studies reported the possibility of using EEG measures to predict response to different pharmacological and non-pharmacological treatments in established TRD. Limitations: The definition of TRD, the selection of variables together with ML algorithms and pipelines varies across the studies, ultimately determining the unfeasibility to implement these models in clinical practice. Conclusions: The findings suggest that ML could be a valid approach to increase our understanding of TRD and to better classify and stratify this disorder, which may ultimately help clinicians in the assessment of major depressive disorders.",
keywords = "Depression, EEG, Imaging, Machine Learning, Major depressive disorder, Treatment resistant depression",
author = "Alessandro Pigoni and Giuseppe Delvecchio and Domenico Madonna and Cinzia Bressi and Jair Soares and Paolo Brambilla",
year = "2019",
month = "12",
day = "1",
doi = "10.1016/j.jad.2019.08.009",
language = "English",
volume = "259",
pages = "21--26",
journal = "Journal of Affective Disorders",
issn = "0165-0327",
publisher = "Elsevier",

}

TY - JOUR

T1 - Can Machine Learning help us in dealing with treatment resistant depression? A review

AU - Pigoni, Alessandro

AU - Delvecchio, Giuseppe

AU - Madonna, Domenico

AU - Bressi, Cinzia

AU - Soares, Jair

AU - Brambilla, Paolo

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Background: About one third of patients treated with antidepressant do not show sufficient symptoms relief and up to 15% of patients remain symptomatic even after multiple trials are applied, configuring a state called treatment resistant depression (TRD). A clear definition of this state and the understanding of underlying mechanisms contributing to chronic disability caused by major depressive disorder is still unknown. Therefore, Machine Learning (ML) techniques emerged in the last years as interesting approaches to deal with such complex problems. Methods: We performed a bibliographic search on Pubmed, Google Scholar and Medline of clinical, imaging, genetic and EEG ML classification studies on treatment-responding depression and TRD as well as studies trying to predict response to a specific treatment in already established TRD. The inclusion criteria were met by eleven studies. Seven focused on the definition of predictors of TRD onset while four attempted to predict the response to specific treatments in TRD. Results: The results showed that it seems possible to classify between responders MDD and TRD with good accuracies based on clinical variables. Moreover, some studies reported the possibility of using EEG measures to predict response to different pharmacological and non-pharmacological treatments in established TRD. Limitations: The definition of TRD, the selection of variables together with ML algorithms and pipelines varies across the studies, ultimately determining the unfeasibility to implement these models in clinical practice. Conclusions: The findings suggest that ML could be a valid approach to increase our understanding of TRD and to better classify and stratify this disorder, which may ultimately help clinicians in the assessment of major depressive disorders.

AB - Background: About one third of patients treated with antidepressant do not show sufficient symptoms relief and up to 15% of patients remain symptomatic even after multiple trials are applied, configuring a state called treatment resistant depression (TRD). A clear definition of this state and the understanding of underlying mechanisms contributing to chronic disability caused by major depressive disorder is still unknown. Therefore, Machine Learning (ML) techniques emerged in the last years as interesting approaches to deal with such complex problems. Methods: We performed a bibliographic search on Pubmed, Google Scholar and Medline of clinical, imaging, genetic and EEG ML classification studies on treatment-responding depression and TRD as well as studies trying to predict response to a specific treatment in already established TRD. The inclusion criteria were met by eleven studies. Seven focused on the definition of predictors of TRD onset while four attempted to predict the response to specific treatments in TRD. Results: The results showed that it seems possible to classify between responders MDD and TRD with good accuracies based on clinical variables. Moreover, some studies reported the possibility of using EEG measures to predict response to different pharmacological and non-pharmacological treatments in established TRD. Limitations: The definition of TRD, the selection of variables together with ML algorithms and pipelines varies across the studies, ultimately determining the unfeasibility to implement these models in clinical practice. Conclusions: The findings suggest that ML could be a valid approach to increase our understanding of TRD and to better classify and stratify this disorder, which may ultimately help clinicians in the assessment of major depressive disorders.

KW - Depression

KW - EEG

KW - Imaging

KW - Machine Learning

KW - Major depressive disorder

KW - Treatment resistant depression

UR - http://www.scopus.com/inward/record.url?scp=85070674339&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070674339&partnerID=8YFLogxK

U2 - 10.1016/j.jad.2019.08.009

DO - 10.1016/j.jad.2019.08.009

M3 - Article

AN - SCOPUS:85070674339

VL - 259

SP - 21

EP - 26

JO - Journal of Affective Disorders

JF - Journal of Affective Disorders

SN - 0165-0327

ER -