Predicting short term mood developments among depressed patients using adherence and ecological momentary assessment data

Adam Mikus, Mark Hoogendoorn, Artur Rocha, Joao Gama, Jeroen Ruwaard, Heleen Riper

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Technology driven interventions provide us with an increasing amount of fine-grained data about the patient. This data includes regular ecological momentary assessments (EMA) but also response times to EMA questions by a user. When observing this data, we see a huge variation between the patterns exhibited by different patients. Some are more stable while others vary a lot over time. This poses a challenging problem for the domain of artificial intelligence and makes on wondering whether it is possible to predict the future mental state of a patient using the data that is available. In the end, these predictions could potentially contribute to interventions that tailor the feedback to the user on a daily basis, for example by warning a user that a fall-back might be expected during the next days, or by applying a strategy to prevent the fall-back from occurring in the first place. In this work, we focus on short term mood prediction by considering the adherence and usage data as an additional predictor. We apply recurrent neural networks to handle the temporal aspects best and try to explore whether individual, group level, or one single predictive model provides the highest predictive performance (measured using the root mean squared error (RMSE)). We use data collected from patients from five countries who used the ICT4Depression/MoodBuster platform in the context of the EU E-COMPARED project. In total, we used the data from 143 patients (with between 9 and 425 days of EMA data) who were diagnosed with a major depressive disorder according to DSM-IV. Results show that we can make predictions of short term mood change quite accurate (ranging between 0.065 and 0.11). The past EMA mood ratings proved to be the most influential while adherence and usage data did not improve prediction accuracy. In general, group level predictions proved to be the most promising, however differences were not significant. Short term mood prediction remains a difficult task, but from this research we can conclude that sophisticated machine learning algorithms/setups can result in accurate performance. For future work, we want to use more data from the mobile phone to improve predictive performance of short term mood.

Original languageEnglish
Pages (from-to)105-110
Number of pages6
JournalInternet Interventions
Volume12
Early online date7 Oct 2017
DOIs
Publication statusPublished - Jun 2018

Fingerprint

Patient Compliance
Cell Phones
Artificial Intelligence
Major Depressive Disorder
Diagnostic and Statistical Manual of Mental Disorders
Reaction Time
Ecological Momentary Assessment
Technology
Research

Keywords

  • Depression
  • Machine learning
  • Prediction
  • Short term mood

Cite this

@article{3d8799630d84498186fd7f4c5766daf2,
title = "Predicting short term mood developments among depressed patients using adherence and ecological momentary assessment data",
abstract = "Technology driven interventions provide us with an increasing amount of fine-grained data about the patient. This data includes regular ecological momentary assessments (EMA) but also response times to EMA questions by a user. When observing this data, we see a huge variation between the patterns exhibited by different patients. Some are more stable while others vary a lot over time. This poses a challenging problem for the domain of artificial intelligence and makes on wondering whether it is possible to predict the future mental state of a patient using the data that is available. In the end, these predictions could potentially contribute to interventions that tailor the feedback to the user on a daily basis, for example by warning a user that a fall-back might be expected during the next days, or by applying a strategy to prevent the fall-back from occurring in the first place. In this work, we focus on short term mood prediction by considering the adherence and usage data as an additional predictor. We apply recurrent neural networks to handle the temporal aspects best and try to explore whether individual, group level, or one single predictive model provides the highest predictive performance (measured using the root mean squared error (RMSE)). We use data collected from patients from five countries who used the ICT4Depression/MoodBuster platform in the context of the EU E-COMPARED project. In total, we used the data from 143 patients (with between 9 and 425 days of EMA data) who were diagnosed with a major depressive disorder according to DSM-IV. Results show that we can make predictions of short term mood change quite accurate (ranging between 0.065 and 0.11). The past EMA mood ratings proved to be the most influential while adherence and usage data did not improve prediction accuracy. In general, group level predictions proved to be the most promising, however differences were not significant. Short term mood prediction remains a difficult task, but from this research we can conclude that sophisticated machine learning algorithms/setups can result in accurate performance. For future work, we want to use more data from the mobile phone to improve predictive performance of short term mood.",
keywords = "Depression, Machine learning, Prediction, Short term mood",
author = "Adam Mikus and Mark Hoogendoorn and Artur Rocha and Joao Gama and Jeroen Ruwaard and Heleen Riper",
year = "2018",
month = "6",
doi = "10.1016/j.invent.2017.10.001",
language = "English",
volume = "12",
pages = "105--110",
journal = "Internet Interventions",
issn = "2214-7829",
publisher = "Elsevier BV",

}

Predicting short term mood developments among depressed patients using adherence and ecological momentary assessment data. / Mikus, Adam; Hoogendoorn, Mark; Rocha, Artur; Gama, Joao; Ruwaard, Jeroen; Riper, Heleen.

In: Internet Interventions, Vol. 12, 06.2018, p. 105-110.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Predicting short term mood developments among depressed patients using adherence and ecological momentary assessment data

AU - Mikus, Adam

AU - Hoogendoorn, Mark

AU - Rocha, Artur

AU - Gama, Joao

AU - Ruwaard, Jeroen

AU - Riper, Heleen

PY - 2018/6

Y1 - 2018/6

N2 - Technology driven interventions provide us with an increasing amount of fine-grained data about the patient. This data includes regular ecological momentary assessments (EMA) but also response times to EMA questions by a user. When observing this data, we see a huge variation between the patterns exhibited by different patients. Some are more stable while others vary a lot over time. This poses a challenging problem for the domain of artificial intelligence and makes on wondering whether it is possible to predict the future mental state of a patient using the data that is available. In the end, these predictions could potentially contribute to interventions that tailor the feedback to the user on a daily basis, for example by warning a user that a fall-back might be expected during the next days, or by applying a strategy to prevent the fall-back from occurring in the first place. In this work, we focus on short term mood prediction by considering the adherence and usage data as an additional predictor. We apply recurrent neural networks to handle the temporal aspects best and try to explore whether individual, group level, or one single predictive model provides the highest predictive performance (measured using the root mean squared error (RMSE)). We use data collected from patients from five countries who used the ICT4Depression/MoodBuster platform in the context of the EU E-COMPARED project. In total, we used the data from 143 patients (with between 9 and 425 days of EMA data) who were diagnosed with a major depressive disorder according to DSM-IV. Results show that we can make predictions of short term mood change quite accurate (ranging between 0.065 and 0.11). The past EMA mood ratings proved to be the most influential while adherence and usage data did not improve prediction accuracy. In general, group level predictions proved to be the most promising, however differences were not significant. Short term mood prediction remains a difficult task, but from this research we can conclude that sophisticated machine learning algorithms/setups can result in accurate performance. For future work, we want to use more data from the mobile phone to improve predictive performance of short term mood.

AB - Technology driven interventions provide us with an increasing amount of fine-grained data about the patient. This data includes regular ecological momentary assessments (EMA) but also response times to EMA questions by a user. When observing this data, we see a huge variation between the patterns exhibited by different patients. Some are more stable while others vary a lot over time. This poses a challenging problem for the domain of artificial intelligence and makes on wondering whether it is possible to predict the future mental state of a patient using the data that is available. In the end, these predictions could potentially contribute to interventions that tailor the feedback to the user on a daily basis, for example by warning a user that a fall-back might be expected during the next days, or by applying a strategy to prevent the fall-back from occurring in the first place. In this work, we focus on short term mood prediction by considering the adherence and usage data as an additional predictor. We apply recurrent neural networks to handle the temporal aspects best and try to explore whether individual, group level, or one single predictive model provides the highest predictive performance (measured using the root mean squared error (RMSE)). We use data collected from patients from five countries who used the ICT4Depression/MoodBuster platform in the context of the EU E-COMPARED project. In total, we used the data from 143 patients (with between 9 and 425 days of EMA data) who were diagnosed with a major depressive disorder according to DSM-IV. Results show that we can make predictions of short term mood change quite accurate (ranging between 0.065 and 0.11). The past EMA mood ratings proved to be the most influential while adherence and usage data did not improve prediction accuracy. In general, group level predictions proved to be the most promising, however differences were not significant. Short term mood prediction remains a difficult task, but from this research we can conclude that sophisticated machine learning algorithms/setups can result in accurate performance. For future work, we want to use more data from the mobile phone to improve predictive performance of short term mood.

KW - Depression

KW - Machine learning

KW - Prediction

KW - Short term mood

UR - http://www.scopus.com/inward/record.url?scp=85031319063&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031319063&partnerID=8YFLogxK

U2 - 10.1016/j.invent.2017.10.001

DO - 10.1016/j.invent.2017.10.001

M3 - Article

VL - 12

SP - 105

EP - 110

JO - Internet Interventions

JF - Internet Interventions

SN - 2214-7829

ER -