Predicting Depression in Children and Adolescents using the SHAP Approach Marcelo Balbino 1,2 , Renata Santana 2 , Maycoln Teodoro 3 , Mark Song 2 , Luis Z´ arate 2 and Cristiane Nobre 2 1 Department of Computing and Civil Construction, Federal Center for Technological Education of Minas Gerais, Brazil 2 Department of Computing, Pontifical Catholic University of Minas Gerais University, Brazil 3 Department of Psychology, Federal University of Minas Gerais, Brazil Keywords: Depression, Machine Learning, Interpretability, SHAP. Abstract: Depression is a disease with severe consequences that affects millions of people, with the onset of the first symptoms being common in youth. It is essential to identify and treat individuals with depression as early as possible to prevent the losses caused by the disorder throughout life. However, the diagnostic criteria of depressive disorders for children/adolescents or adults is not differentiated, even though authors claim that the particularities of childhood must be considered. This may be why childhood depression is being underdiag- nosed. Therefore, this work aims to discover the most significant features in diagnosing depression in children and adolescents through Machine Learning methods and the SHAP approach. Models with Machine Learning algorithms were developed, and the model with SVM presented the best results. The application of SHAP proved to be fundamental to deepen the understanding of this model. The experiments indicated that feelings of isolation, sadness, excessive worry, complaints about one’s appearance, resistance to academic tasks, and the mother’s schooling are the most significant features in predicting depression in children and adolescents. Such results can help to understand depression in these individuals and thus lead to appropriate treatment. 1 INTRODUCTION Depression is a term used to refer to Depressive Dis- orders, being understood as a pathology that alters and compromises the body and mind, mainly affect- ing mood. The individual with Depressive Disorders may have persistent sadness, lack of interest or plea- sure in previously rewarding activities, loss of con- fidence and self-esteem, unjustified feelings of guilt, ideas of death and suicide, sleep and appetite distur- bances, fatigue, poor concentration, and symptoms of anxiety. Its effects can be long-lasting or recurrent and can affect a person’s ability in essential areas of functioning (APA et al., 2013; WHO, 2017). From 2005 to 2015, there was an 18% increase in people with depression worldwide, resulting in more than 300 million people (WHO, 2017). Fur- thermore, it is estimated that one in six people (about 16.67%) will suffer from depression at some point in their lives, which means more than one billion people worldwide affected by the disorder (APA, 2017). Studies indicate that Depressive Disorders have been the leading cause of illnesses and disabilities in adolescence (WHO, 2017). In addition, half of the people who develop mental disorders experience the first symptoms by 14 years (Yoon et al., 2014). There- fore, it is essential to identify and treat individuals with depression in childhood/adolescence to prevent the losses caused by the disorder throughout life. The definition of depression in youth is not specif- ically addressed in the Diagnostic and Statistical Manual of Mental Disorders (APA et al., 2013). There is no differentiation of diagnostic criteria for depres- sive disorders for children, adolescents, or adults. Nevertheless, authors claim that the peculiarities of childhood must be considered in the assessment and diagnosis of depression in children (Quevedo et al., 2018; Bernaras et al., 2019). However, one of the obstacles to treating depres- sion is its assessment and diagnosis, leading to a lack of treatment or inadequate handling of it (Pavlova and Uher, 2020). This scenario highlights the importance of instruments that can support the correct diagnosis. A survey gave rise to a database containing infor- mation on 377 children and adolescents with different depressive symptomatology. 514 Balbino, M., Santana, R., Teodoro, M., Song, M., Zárate, L. and Nobre, C. Predicting Depression in Children and Adolescents using the SHAP Approach. DOI: 10.5220/0010842500003123 In Proceedings of the 15th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2022) - Volume 5: HEALTHINF, pages 514-521 ISBN: 978-989-758-552-4; ISSN: 2184-4305 Copyright c  2022 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved