The promise and challenges of computer mouse trajectories in DMHIs – A feasibility study on pre-treatment dropout predictions

Research output: Journal contributionsJournal articlesResearchpeer-review

Standard

The promise and challenges of computer mouse trajectories in DMHIs – A feasibility study on pre-treatment dropout predictions. / Zantvoort, Kirsten; Matthiesen, Jennifer; Bjurner, Pontus et al.
In: Internet Interventions, Vol. 40, No. 40, 100828, 06.2025.

Research output: Journal contributionsJournal articlesResearchpeer-review

Harvard

APA

Vancouver

Bibtex

@article{a14178fdaa2f43a0be35398032860279,
title = "The promise and challenges of computer mouse trajectories in DMHIs – A feasibility study on pre-treatment dropout predictions",
abstract = "With the impetus of Digital Mental Health Interventions (DMHIs), complex data can be leveraged to improve and personalize mental health care. However, most approaches rely on a very limited number of often costly features. Computer mouse trajectories can be unobtrusively and cost-efficiently gathered and seamlessly integrated into current baseline processes. Empirical evidence suggests that mouse movements hold information on user motivation and attention, both valuable aspects otherwise difficult to measure at scale. Further, mouse trajectories can already be collected on pre-treatment questionnaires, making them a promising candidate for early predictions informing treatment allocation. Therefore, this paper discusses how to collect and process mouse trajectory data on questionnaires in DMHIs. Covering different complexity levels, we combine hand-crafted features with non-sequential machine learning models, as well as spatiotemporal raw mouse data with state-of-the-art sequential neural networks. The data processing pipeline for the latter includes task-specific pre-processing to convert the variable length trajectories into a single prediction per user. As a feasibility study, we collected mouse trajectory data from 183 patients filling out a pre-intervention depression questionnaire. While the hand-crafted features slightly improve baseline predictions, the spatiotemporal models underperform. However, considering our small data set size, we propose more research to investigate the potential value of this novel and promising data type and provide the necessary steps and open-source code to do so.",
keywords = "Dropout, E-mental health, ICBT, Machine learning, Prediction",
author = "Kirsten Zantvoort and Jennifer Matthiesen and Pontus Bjurner and Marie Bendix and Ulf Brefeld and Burkhardt Funk and Viktor Kaldo",
note = "Publisher Copyright: {\textcopyright} 2025",
year = "2025",
month = jun,
doi = "10.1016/j.invent.2025.100828",
language = "English",
volume = "40",
journal = "Internet Interventions",
issn = "2214-7829",
publisher = "Elsevier B.V.",
number = "40",

}

RIS

TY - JOUR

T1 - The promise and challenges of computer mouse trajectories in DMHIs – A feasibility study on pre-treatment dropout predictions

AU - Zantvoort, Kirsten

AU - Matthiesen, Jennifer

AU - Bjurner, Pontus

AU - Bendix, Marie

AU - Brefeld, Ulf

AU - Funk, Burkhardt

AU - Kaldo, Viktor

N1 - Publisher Copyright: © 2025

PY - 2025/6

Y1 - 2025/6

N2 - With the impetus of Digital Mental Health Interventions (DMHIs), complex data can be leveraged to improve and personalize mental health care. However, most approaches rely on a very limited number of often costly features. Computer mouse trajectories can be unobtrusively and cost-efficiently gathered and seamlessly integrated into current baseline processes. Empirical evidence suggests that mouse movements hold information on user motivation and attention, both valuable aspects otherwise difficult to measure at scale. Further, mouse trajectories can already be collected on pre-treatment questionnaires, making them a promising candidate for early predictions informing treatment allocation. Therefore, this paper discusses how to collect and process mouse trajectory data on questionnaires in DMHIs. Covering different complexity levels, we combine hand-crafted features with non-sequential machine learning models, as well as spatiotemporal raw mouse data with state-of-the-art sequential neural networks. The data processing pipeline for the latter includes task-specific pre-processing to convert the variable length trajectories into a single prediction per user. As a feasibility study, we collected mouse trajectory data from 183 patients filling out a pre-intervention depression questionnaire. While the hand-crafted features slightly improve baseline predictions, the spatiotemporal models underperform. However, considering our small data set size, we propose more research to investigate the potential value of this novel and promising data type and provide the necessary steps and open-source code to do so.

AB - With the impetus of Digital Mental Health Interventions (DMHIs), complex data can be leveraged to improve and personalize mental health care. However, most approaches rely on a very limited number of often costly features. Computer mouse trajectories can be unobtrusively and cost-efficiently gathered and seamlessly integrated into current baseline processes. Empirical evidence suggests that mouse movements hold information on user motivation and attention, both valuable aspects otherwise difficult to measure at scale. Further, mouse trajectories can already be collected on pre-treatment questionnaires, making them a promising candidate for early predictions informing treatment allocation. Therefore, this paper discusses how to collect and process mouse trajectory data on questionnaires in DMHIs. Covering different complexity levels, we combine hand-crafted features with non-sequential machine learning models, as well as spatiotemporal raw mouse data with state-of-the-art sequential neural networks. The data processing pipeline for the latter includes task-specific pre-processing to convert the variable length trajectories into a single prediction per user. As a feasibility study, we collected mouse trajectory data from 183 patients filling out a pre-intervention depression questionnaire. While the hand-crafted features slightly improve baseline predictions, the spatiotemporal models underperform. However, considering our small data set size, we propose more research to investigate the potential value of this novel and promising data type and provide the necessary steps and open-source code to do so.

KW - Dropout

KW - E-mental health

KW - ICBT

KW - Machine learning

KW - Prediction

UR - http://www.scopus.com/inward/record.url?scp=105002214015&partnerID=8YFLogxK

UR - https://www.sciencedirect.com/journal/internet-interventions/vol/40/suppl/C

UR - https://www.sciencedirect.com/science/article/pii/S2214782925000296

U2 - 10.1016/j.invent.2025.100828

DO - 10.1016/j.invent.2025.100828

M3 - Journal articles

VL - 40

JO - Internet Interventions

JF - Internet Interventions

SN - 2214-7829

IS - 40

M1 - 100828

ER -

Recently viewed

Publications

  1. On the combined effect of soil fertility and topography on tree growth in subtropical forest ecosystems - a study from SE China
  2. Control of Permanent Magnet Synchronous Motors for Track Applications
  3. Anonymity reprogrammed
  4. Identity construction and representation in education - centred internet memes
  5. Non-linear effects of comparison income in quit decisions: status versus signal !
  6. Increasing the acceptance of internet-based mental health interventions in primary care patients with depressive symptoms
  7. Empathy and empathic leadership practices in schools – a scoping review
  8. Can adults learn L2 grammar after prolonged exposure under incidental conditions?
  9. Analyzing pre- and in-service teachers’ feedback practice with microteaching videos
  10. Applying the principles of green engineering to cradle-to-cradle design
  11. LiteraturGesellschaft DDR
  12. Managing and accounting for corporate biodiversity contributions mapping the field
  13. First automatic size measurements for the separation of dwarf birch and tree birch pollen in MIS 6 to MIS 1 records from Northern Germany
  14. Social network activities as a predictor for phase transitions of patients with bipolar affective disorders
  15. Effect of cascading of higher-lying states on a delayed 1 s-2 p transition after beam-foil excitation of 56 MeV hydrogen-like oxygen and fluorine
  16. where paintings live
  17. A Process Perspective on Organizational Failure
  18. Internet- and App-Based Stress Intervention for Distance-Learning Students With Depressive Symptoms
  19. Über Franz Hessel
  20. Probing alignment of personal and organisational values for sustainability
  21. Introduction - How prenatal diagnosis is entangled in historical and social contexts
  22. Citizen Relationship Management (CRM)
  23. Credit constraints, endogenous innovations, and price setting in international trade
  24. Effect of laser peen forming process parameters on bending and surface quality of Ti-6Al-4V sheets
  25. Non-native Douglas fir promotes epigeal spider density, but has a mixed effect on functional diversity
  26. Effects on the (CSR) Reputation
  27. Grausamer Optimismus
  28. The untapped potential of Games for Health in times of crises. A critical reflection
  29. Skill learning as a concept in life-span developmental psychology
  30. Minimalist Training
  31. Hidden in full view
  32. Note on the Lingen case
  33. Analysis of Kinetic Dynamics of the Multipole Resonance Probe
  34. Academia's obsession with quantity
  35. Behind the Scenes of Automation
  36. Community resilience for a 1.5 degrees C world