Improving Web-Based Treatment Intake for Multiple Mental and Substance Use Disorders by Text Mining and Machine Learning: Algorithm Development and Validation.
automated intake and referral
computerized CBT
mental health disorders
multi-class classification
screening
supervised text classification
Journal
JMIR mental health
ISSN: 2368-7959
Titre abrégé: JMIR Ment Health
Pays: Canada
ID NLM: 101658926
Informations de publication
Date de publication:
11 Apr 2022
11 Apr 2022
Historique:
received:
05
06
2020
accepted:
28
09
2021
revised:
01
11
2020
entrez:
11
4
2022
pubmed:
12
4
2022
medline:
12
4
2022
Statut:
epublish
Résumé
Text mining and machine learning are increasingly used in mental health care practice and research, potentially saving time and effort in the diagnosis and monitoring of patients. Previous studies showed that mental disorders can be detected based on text, but they focused on screening for a single predefined disorder instead of multiple disorders simultaneously. The aim of this study is to develop a Dutch multi-class text-classification model to screen for a range of mental disorders to refer new patients to the most suitable treatment. On the basis of textual responses of patients (N=5863) to a questionnaire currently used for intake and referral, a 7-class classifier was developed to distinguish among anxiety, panic, posttraumatic stress, mood, eating, substance use, and somatic symptom disorders. A linear support vector machine was fitted using nested cross-validation grid search. The highest classification rate was found for eating disorders (82%). The scores for panic (55%), posttraumatic stress (52%), mood (50%), somatic symptom (50%), anxiety (35%), and substance use disorders (33%) were lower, likely because of overlapping symptoms. The overall classification accuracy (49%) was reasonable for a 7-class classifier. A classification model was developed that could screen text for multiple mental health disorders. The screener resulted in an additional outcome score that may serve as input for a formal diagnostic interview and referral. This may lead to a more efficient and standardized intake process.
Sections du résumé
BACKGROUND
BACKGROUND
Text mining and machine learning are increasingly used in mental health care practice and research, potentially saving time and effort in the diagnosis and monitoring of patients. Previous studies showed that mental disorders can be detected based on text, but they focused on screening for a single predefined disorder instead of multiple disorders simultaneously.
OBJECTIVE
OBJECTIVE
The aim of this study is to develop a Dutch multi-class text-classification model to screen for a range of mental disorders to refer new patients to the most suitable treatment.
METHODS
METHODS
On the basis of textual responses of patients (N=5863) to a questionnaire currently used for intake and referral, a 7-class classifier was developed to distinguish among anxiety, panic, posttraumatic stress, mood, eating, substance use, and somatic symptom disorders. A linear support vector machine was fitted using nested cross-validation grid search.
RESULTS
RESULTS
The highest classification rate was found for eating disorders (82%). The scores for panic (55%), posttraumatic stress (52%), mood (50%), somatic symptom (50%), anxiety (35%), and substance use disorders (33%) were lower, likely because of overlapping symptoms. The overall classification accuracy (49%) was reasonable for a 7-class classifier.
CONCLUSIONS
CONCLUSIONS
A classification model was developed that could screen text for multiple mental health disorders. The screener resulted in an additional outcome score that may serve as input for a formal diagnostic interview and referral. This may lead to a more efficient and standardized intake process.
Identifiants
pubmed: 35404261
pii: v9i4e21111
doi: 10.2196/21111
pmc: PMC9039807
doi:
Types de publication
Journal Article
Langues
eng
Pagination
e21111Informations de copyright
©Sytske Wiegersma, Maurice Hidajat, Bart Schrieken, Bernard Veldkamp, Miranda Olff. Originally published in JMIR Mental Health (https://mental.jmir.org), 11.04.2022.
Références
Assessment. 2017 Mar;24(2):157-172
pubmed: 26358713
BMC Psychiatry. 2006 Aug 22;6:34
pubmed: 16925825
Psychol Med. 2010 Dec;40(12):1943-57
pubmed: 20406528
Artif Intell Med. 2012 Sep;56(1):19-25
pubmed: 22771201
J Med Internet Res. 2017 Jun 29;19(6):e228
pubmed: 28663166
Psychother Psychosom. 2005;74(6):336-43
pubmed: 16244509
Behav Res Ther. 2011 Mar;49(3):158-69
pubmed: 21255767
BMC Psychiatry. 2017 Dec 01;17(1):382
pubmed: 29191173
J Med Internet Res. 2013 Nov 01;15(11):e239
pubmed: 24184993
Eur J Psychotraumatol. 2015 May 19;6:27882
pubmed: 25994025
Eur J Psychotraumatol. 2018 Feb 06;9(sup1):1424448
pubmed: 29441154
Eur J Psychotraumatol. 2017 Oct 09;8(1):1375338
pubmed: 29435198
Clin Psychol Rev. 2004 Sep;24(5):583-616
pubmed: 15325746
Psychol Med. 2007 Mar;37(3):319-28
pubmed: 17112400
Bull World Health Organ. 2004 Nov;82(11):858-66
pubmed: 15640922
Comput Methods Programs Biomed. 2012 Mar;105(3):194-209
pubmed: 22070853
Arch Gen Psychiatry. 1995 Dec;52(12):1048-60
pubmed: 7492257
Eur J Psychotraumatol. 2018 Mar 8;8(Suppl 5):1441582
pubmed: 33178404
Lancet. 2016 Oct 8;388(10053):1545-1602
pubmed: 27733282
World Psychiatry. 2014 Oct;13(3):306-9
pubmed: 25273304
Int J Eat Disord. 2005;37 Suppl:S80-6; discussion S87-9
pubmed: 15852327
Nature. 2011 Jul 06;475(7354):27-30
pubmed: 21734685
PLoS One. 2012;7(7):e40089
pubmed: 22792217
Int J Methods Psychiatr Res. 2016 Jun;25(2):86-100
pubmed: 26184780
Health Technol Assess. 2006 Sep;10(33):iii, xi-xiv, 1-168
pubmed: 16959169
JAMA. 1990 Nov 21;264(19):2511-8
pubmed: 2232018
J Clin Psychiatry. 2006 Feb;67(2):247-57
pubmed: 16566620
Theor Med Bioeth. 2015 Feb;36(1):41-60
pubmed: 25636962
Cogn Behav Ther. 2011;40(4):251-66
pubmed: 22060248
Drug Alcohol Depend. 2011 Jan 15;113(2-3):147-56
pubmed: 20801586
Psychol Sci. 2003 Jan;14(1):60-5
pubmed: 12564755