Skip to main content

STUDY PROTOCOL article

Front. Psychiatry
Sec. Digital Mental Health
Volume 15 - 2024 | doi: 10.3389/fpsyt.2024.1465933
This article is part of the Research Topic Smart Sensing and Processing for Intelligent Mental Health Detection View all 4 articles

Real-time Monitoring to Predict Depressive Symptoms: study protocol

Provisionally accepted
JongSun Lee JongSun Lee *YuRim Lee YuRim Lee
  • Kangwon National University, Chuncheon, Republic of Korea

The final, formatted version of the article will be published soon.

    Introduction: According to the World Health Organization, Depression is the fourth leading cause of global disease burden. However, traditional clinical and self-report assessments of depression have limitations in providing timely diagnosis and intervention. Recently, digital phenotyping studies have found the possibility of overcoming these limitations through the use of wearable-devices and smartphones. The present study aims to identify the digital phenotype that significantly predicts depressive symptoms. Methods and Analysis: The study will recruit a total of 150 participants in their 20s who have experienced depression for the past two weeks in Korea. The study will collect passive (eg., active energy, exercise minutes, heart rate, heart rate variability, resting energy, resting heart rate, sleep patterns, steps, walking pace) data and Ecological Momentary Assessment (EMA) through smartphone and wearable-device for two weeks. This study will be conducted longitudinally, with two repeated measurements over three months. Passive data will be collected through sensors on the wearable-device, while EMA data will be collected four times a day through a smartphone app. A machine learning algorithm and multilevel model will be used to construct a predictive model for depressive symptoms using the collected data. Discussion: This study explores the potential of wearable devices and smartphones to improve the understanding and treatment of depression in young adults. By collecting continuous, real-time data on physiological and behavioral patterns, the research uncovers subtle changes in heart rate, activity levels and sleep that correlate with depressive symptoms, providing a deeper understanding of the disorder. The findings provide a foundation for further research and contribute to the advancement of digital mental health. Advances in these areas of research may have implications for the detection and prevention of early warning signs of depression through the use of digital markers.

    Keywords: digital phenotyping1, depressive disorder2, ecological momentary assessment3, wearable device4, multilevel modeling5, machine learning6

    Received: 17 Jul 2024; Accepted: 30 Dec 2024.

    Copyright: © 2024 Lee and Lee. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: JongSun Lee, Kangwon National University, Chuncheon, Republic of Korea

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.