Prediction of Adolescent Depression from Prenatal and Childhood Data from the Avon Longitudinal Study of Parents and Children (ALSPAC) Using Machine Learning

Aug 6, 2026

Upload by Hazel

Depression is a leading contributor to disability and mortality for young people and is often first diagnosed in adolescence. This thesis presents a machine learning framework to predict adolescent depression between ages 12 and 18 using environmental, biological, and lifestyle features from the child, mother, and partner, drawn from prenatal life through age 10 within the Avon Longitudinal Study of Parents and Children (ALSPAC). Models were trained and compared across cross-sectional and longitudinal methods, achieving recall 0.59 ± 0.20, specificity 0.61 ± 0.17, and accuracy 0.64 ± 0.13. The most informative features included sex, parental depression and anxiety, and exposure to stressful events or environments, supporting early, evidence-driven risk decision support for mental illness prevention.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages102

Reading Time4 h 17 min

AccessFree to download

Introduction

Methods

Dataset

Sample description

Depression variable description

Results

UC Davis UC Davis Electronic Theses and Dissertations Title Prediction of Adolescent Depression from Prenatal and Childhood Data from the Avon Longitudinal Study of Parents and Children (ALSPAC) Using Machine Learning Permalink [https://escholarship.org/uc/item/8fv3f09r](https://escholarship.org/uc/item/8fv3f09r) ISBN 9798290647340 Author Yoo, Arielle Soomi Publication Date 2025-06-30 Peer reviewed [eScholarship.org](eScholarship.org) Powered by the California Digital Library University of California Prediction of Adolescent Depression from Prenatal and Childhood Data from the Avon Longitudinal Study of Parents and Children (ALSPAC) Using Machine Learning By ARIELLE SOOMI YOO THESIS Submitted in partial satisfaction of the requirements for the degree of MASTER OF SCIENCE in Biomedical Engineering in the OFFICE OF GRADUATE STUDIES of the UNIVERSITY OF CALIFORNIA DAVIS Approved: | Ilias Tagkopoulos, Chair | | --- | | Justin Siegel | Camelia E. Hostinar Committee in Charge 2025 Acknowledgments We are extremely grateful to all the families who took part in this study, the midwives for their help in recruiting them, and the whole ALSPAC team, which includes interviewers, computer and laboratory technicians, clerical workers, research scientists, volunteers, managers, receptionists, and nurses. The UK Medical Research Council, Wellcome (Grant ref: 217065/Z/19/Z), and the University of Bristol provide core support for ALSPAC. A comprehensive list of grant funding is available on the ALSPAC website ( [http://www.bristol.ac.uk/alspac/external/documents/grant-acknowledgements.pdf](http://www.bristol.ac.uk/alspac/external/documents/grant-acknowledgements.pdf)). This research was speciﬁcally funded by MRC Grant G0401540 73080. This publication is the work of the authors (Arielle Yoo, Fangzhou Li, Jason Youn, Joanna Guan, Amanda Guyer, Camelia Hostinar, and Ilias Tagkopoulos), who will serve as guarantors for the contents of this paper. Amanda Guyer, Camelia Hostinar, and Ilias Tagkopoulos were supported by R21MH125346 . We would also like to thank LillyBelle Deer, Ph.D., for her assistance with the initial stages of this project. Permissions/Acknowledgements for use of Copyrighted Materials Much of this work was published previously in Scientiﬁc Reports (Yoo, et al, 2024)1 under Creative Commons Attribution 4.0 International License2 . This work has been a collaborative efort by Arielle Yoo, Fangzhou Li, Jason Youn, Joanna Guan, Amanda Guyer, Camelia Hostinar, and Ilias Tagkopoulos. I have performed all computational analyses presented, and have created all ﬁgures unless otherwise noted. I have written the original manuscript draft of the published paper, which is the basis of this master thesis. However, that work has been substantially amended, edited, and revised by my co-authors, so authorship of the text that has been copied from that work is jointly shared among all the co-authors. Abstract Depression is a major cause of disability and mortality for young people worldwide and is typically ﬁrst diagnosed during adolescence. In this work, we present a machine learning framework to predict adolescent depression occurring between ages 12 and 18 years using environmental, biological, and lifestyle features of the child, mother, and partner from the child’s prenatal period to age 10 years using data from 8,467 participants enrolled in the Avon Longitudinal Study of Parents and Children (ALSPAC) . We trained and compared several cross-sectional and longitudinal machine learning techniques and found the resulting models predicted adolescent depression with recall (0.59 ± 0.20), speciﬁcity (0.61 ± 0.17), and accuracy (0.64 ± 0.13), using on average 39 out of the 885 total features (4.4%) included in the models. The leading informative features in our predictive models of adolescent depression were sex, parental depression and anxiety, and exposure to st

FAQs

What prediction task does this thesis address?

It predicts adolescent depression occurring between ages 12 and 18.

Which data sources and time window are used for prediction?

It uses environmental, biological, and lifestyle features from the child, mother, and partner from the child’s prenatal period to age 10, based on ALSPAC data.

Which modeling approaches and performance results are reported?

The work trains and compares cross-sectional and longitudinal machine learning techniques, reporting recall 0.59 ± 0.20, specificity 0.61 ± 0.17, and accuracy 0.64 ± 0.13.

Aug 6, 2026

Upload by Hazel