Use of Machine-Learning Algorithms Based on Text, Audio and Video Data in the Prediction of Anxiety and Post-Traumatic Stress in General and Clinical Populations: A Systematic Review

Research output: Journal contributionsScientific review articlesResearch

Authors

  • Marketa Ciharova
  • Khadicha Amarti
  • Ward van Breda
  • Xianhua Peng
  • Rosa Lorente-Català
  • Burkhardt Funk
  • Mark Hoogendoorn
  • Nikolaos Koutsouleris
  • Paolo Fusar-Poli
  • Eirini Karyotaki
  • Pim Cuijpers
  • Heleen Riper

Research in machine learning (ML) algorithms using natural behavior (i.e., text, audio, and video data) suggests that these techniques could contribute to personalization in psychology and psychiatry. However, a systematic review of the current state of the art is missing. Moreover, individual studies often target ML experts who may overlook potential clinical implications of their findings. In a narrative accessible to mental health professionals, we present a systematic review conducted in 5 psychology and 2 computer science databases. We included 128 studies that assessed the predictive power of ML algorithms using text, audio, and/or video data in the prediction of anxiety and posttraumatic stress disorder. Most studies (n = 87) were aimed at predicting anxiety, while the remainder (n = 41) focused on posttraumatic stress disorder. They were mostly published since 2019 in computer science journals and tested algorithms using text (n = 72) as opposed to audio or video. Studies focused mainly on general populations (n = 92) and less on laboratory experiments (n = 23) or clinical populations (n = 13). Methodological quality varied, as did reported metrics of the predictive power, hampering comparison across studies. Two-thirds of studies, which focused on both disorders, reported acceptable to very good predictive power (including high-quality studies only). The results of 33 studies were uninterpretable, mainly due to missing information. Research into ML algorithms using natural behavior is in its infancy but shows potential to contribute to diagnostics of mental disorders, such as anxiety and posttraumatic stress disorder, in the future if standardization of methods, reporting of results, and research in clinical populations are improved.

Original languageEnglish
JournalBiological Psychiatry
Volume96
Issue number7
Pages (from-to)519-531
Number of pages13
ISSN0006-3223
DOIs
Publication statusPublished - 01.10.2024

Bibliographical note

Publisher Copyright:
© 2024 Society of Biological Psychiatry

    Research areas

  • Informatics - Anxiety, Audio, Machine Learning, Posttraumatic stress, Text, Video

Recently viewed

Publications

  1. On finding nonisomorphic connected subgraphs and distinct molecular substructures.
  2. Changes of Perception
  3. Advantages and Disadvanteges of Different Text Coding Procedures for Research and Practice in a School Context
  4. Geodesign as a boundary management process
  5. Bayesian Analysis of Longitudinal Multitrait
  6. Discussion report part 2
  7. Differentiating forest types using TerraSAR–X spotlight images based on inferential statistics and multivariate analysis
  8. Development and prospects of degradable magnesium alloys for structural and functional applications in the fields of environment and energy
  9. Integrating inductive and deductive analysis to identify and characterize archetypical social-ecological systems and their changes
  10. Feel the Music! Exploring the Cross-modal Correspondence between Music and Haptic Perceptions of Softness
  11. Counteracting electric vehicle range concern with a scalable behavioural intervention
  12. Third International Mathematics and Science Study and Trends in Mathematics and Science Studies (TIMSS)
  13. Public perceptions of CCS
  14. Revisiting Carbon Disclosure and Performance
  15. Efficacy of trapping techniques (pitfall, ramp and arboreal traps) for capturing spiders
  16. E-collaborative knowledge construction in chat environments
  17. Contributing to sustainable development pathways in the South Pacific through transdisciplinary research
  18. Polarization of Time and Income
  19. Identity without Membership?
  20. Net deferred tax assets and the long-run performance of initial public offerings
  21. General strategies to increase the repeatability in non-target screening by liquid chromatography-high resolution mass spectrometry
  22. Do they really care about targeted political ads? Investigation of user privacy concerns and preferences
  23. Introduction
  24. Discussion report part 3
  25. Crowdfunding the Commons?
  26. Paar normal oder paranormal
  27. Periodizing Latin American art since the 1960s
  28. Responsible Research is also concerned with generalizability
  29. "The spirit of Europe" differential migration, labour and logistification