Assessment and Prediction of Depression and Anxiety Risk Factors in Schoolchildren: Machine Learning Techniques Performance Analysis

Radwan Qasrawi; Stephanny Paola Vicuna Polo; Diala Abu Al-Halawa; Sameh Hallaq; Ziad Abdeen

doi:10.2196/32736

Assessment and Prediction of Depression and Anxiety Risk Factors in Schoolchildren: Machine Learning Techniques Performance Analysis

JMIR Form Res. 2022 Aug 31;6(8):e32736. doi: 10.2196/32736.

Authors

Radwan Qasrawi^{1

2}, Stephanny Paola Vicuna Polo³, Diala Abu Al-Halawa⁴, Sameh Hallaq⁵, Ziad Abdeen⁴

Affiliations

¹ Department of Computer Science, Al-Quds University, Ramallah, Occupied Palestinian Territory.
² Department of Computer Engineering, Istinye University, Istanbul, Turkey.
³ Center for Business Innovation and Technology, Al-Quds University, Jerusalem, Occupied Palestinian Territory.
⁴ Faculty of Medicine, Al-Quds University, Jerusalem, Occupied Palestinian Territory.
⁵ Al-Quds Bard College for Arts and Sciences, Al-Quds University, Jerusalem, Occupied Palestinian Territory.

PMID: 35665695
PMCID: PMC9475423
DOI: 10.2196/32736

Abstract

Background: Depression and anxiety symptoms in early childhood have a major effect on children's mental health growth and cognitive development. The effect of mental health problems on cognitive development has been studied by researchers for the last 2 decades.

Objective: In this paper, we sought to use machine learning techniques to predict the risk factors associated with schoolchildren's depression and anxiety.

Methods: The study sample consisted of 3984 students in fifth to ninth grades, aged 10-15 years, studying at public and refugee schools in the West Bank. The data were collected using the health behaviors schoolchildren questionnaire in the 2013-2014 academic year and analyzed using machine learning to predict the risk factors associated with student mental health symptoms. We used 5 machine learning techniques (random forest [RF], neural network, decision tree, support vector machine [SVM], and naive Bayes) for prediction.

Results: The results indicated that the SVM and RF models had the highest accuracy levels for depression (SVM: 92.5%; RF: 76.4%) and anxiety (SVM: 92.4%; RF: 78.6%). Thus, the SVM and RF models had the best performance in classifying and predicting the students' depression and anxiety. The results showed that school violence and bullying, home violence, academic performance, and family income were the most important factors affecting the depression and anxiety scales.

Conclusions: Overall, machine learning proved to be an efficient tool for identifying and predicting the associated factors that influence student depression and anxiety. The machine learning techniques seem to be a good model for predicting abnormal depression and anxiety symptoms among schoolchildren, so the deployment of machine learning within the school information systems might facilitate the development of health prevention and intervention programs that will enhance students' mental health and cognitive development.

Keywords: anxiety; children; depression; early childhood education; machine learning; prediction; random forest; school-age children; schoolchildren; transition-aged youth; young adult; youth.

©Radwan Qasrawi, Stephanny Paola Vicuna Polo, Diala Abu Al-Halawa, Sameh Hallaq, Ziad Abdeen. Originally published in JMIR Formative Research (https://formative.jmir.org), 31.08.2022.