Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: JMIR mHealth and uHealth

Date Submitted: Sep 16, 2020
Date Accepted: Jan 5, 2021

The final, peer-reviewed published version of this preprint can be found here:

Tracking and Monitoring Mood Stability of Patients With Major Depressive Disorder by Machine Learning Models Using Passive Digital Data: Prospective Naturalistic Multicenter Study

Bai R, Xiao L, Guo Y, Zhu X, Li N, Wang Y, Feng L, Chen Q, Wang Y, Yu X, Xie H, Wang G

Tracking and Monitoring Mood Stability of Patients With Major Depressive Disorder by Machine Learning Models Using Passive Digital Data: Prospective Naturalistic Multicenter Study

JMIR Mhealth Uhealth 2021;9(3):e24365

DOI: 10.2196/24365

PMID: 33683207

PMCID: 7985800

Tracking and Monitoring Mood Stability of Patients with Major Depressive Disorder by Machine Learning Models Using Passive Digital Data

  • Ran Bai; 
  • Le Xiao; 
  • Yu Guo; 
  • Xuequan Zhu; 
  • Nanxi Li; 
  • Yashen Wang; 
  • Lei Feng; 
  • Qinqin Chen; 
  • Yinghua Wang; 
  • Xiangyi Yu; 
  • Haiyong Xie; 
  • Gang Wang

ABSTRACT

Background:

Depression is a common mental illness characterized by persistent sadness and a loss of interest in activities. Using smartphones and wearable devices to monitor MDD patients’ mental condition has been examined in several studies. However, there are few studies utilizing passively collected data to monitor mood changes in a time period.

Objective:

We aimed to examine the feasibility of monitoring mood status and stability of MDD patients using machine learning models trained by passively collected data including phone usage data, sleep data and step count data.

Methods:

We constructed 612 data samples representing time spans during 3 consecutive PHQ-9 assessments. Each data sample was labeled as Steady or Mood Swing with subgroups Steady-remission, Steady-depressed, Mood Swing-drastic, Mood Swing-moderate based on patients’ PHQ-9 scores from 3 visits. 252 features were extracted, and 4 feature selection models were applied. 6 different combinations of types of data were experimented using 6 different machine learning models.

Results:

A total of 334 participants with MDD were enrolled in this study. The highest accuracy of classification between Steady and Mood Swing was 76.62% and recall was 91.53% with features from all types of data being used. Among 6 combinations of types of data we experimented, the overall best combination was using Call Logs, Sleep data, Step count data and Heart rate data. The accuracies of predicting between Steady-remission and Mood Swing-drastic, Steady-remission and Mood Swing-moderate, Steady-depressed and Mood Swing-drastic were over 80%, and the accuracy of predicting between Steady-depressed and Mood Swing-moderate and the overall Steady to Mood Swing classification accuracy were over 75%. Comparing all 6 combinations above, we found that the overall prediction accuracies between Steady-remission and Mood Swing (drastic and moderate), are better than those between Steady-depressed and Mood Swing (drastic and moderate).

Conclusions:

Our proposed method could be used to monitor MDD patients’ mood changes with a promising accuracy utilizing passively collected data, which can be used as a reference to doctors for adjusting treatment plans or a warning of relapse to patients and their guardians. Clinical Trial: ChiCTR1900021461


 Citation

Please cite as:

Bai R, Xiao L, Guo Y, Zhu X, Li N, Wang Y, Feng L, Chen Q, Wang Y, Yu X, Xie H, Wang G

Tracking and Monitoring Mood Stability of Patients With Major Depressive Disorder by Machine Learning Models Using Passive Digital Data: Prospective Naturalistic Multicenter Study

JMIR Mhealth Uhealth 2021;9(3):e24365

DOI: 10.2196/24365

PMID: 33683207

PMCID: 7985800

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.