Publication Details

Continuous Assessment of Function and Disability via Mobile Sensing: Real-World Data-Driven Feasibility Study

SUKEI, E.; ROMERO-MEDRANO, L.; DE LEON MARTINEZ, S.; HERRERA, J.; CAMPANA-MONTES, J.; OLMOS, M.; BACA-GARCIA, E.; ARTES, A. Continuous Assessment of Function and Disability via Mobile Sensing: Real-World Data-Driven Feasibility Study. JMIR Formative Research, 2023, vol. 7, no. 2023, p. 1-10. ISSN: 2561-326X.
Czech title
Průběžné hodnocení funkce a postižení prostřednictvím mobilního telefonu Snímání: Studie proveditelnosti založená na datech z reálného světa
Type
journal article
Language
English
Authors
SUKEI, E.
ROMERO-MEDRANO, L.
DE LEON MARTINEZ, S.
HERRERA, J.
CAMPANA-MONTES, J.
OLMOS, M.
BACA-GARCIA, E.
ARTES, A.
URL
Keywords

WHODAS; functional limitations; mobile sensing; passive ecological momentary assessment; predictive modeling; interpretable
machine learning; machine learning; disability; clinical outcome 

Abstract

Background: Functional limitations are associated with poor clinical outcomes, higher mortality, and disability rates, especially
in older adults. Continuous assessment of patients' functionality is important for clinical practice; however, traditional
questionnaire-based assessment methods are very time-consuming and infrequently used. Mobile sensing offers a great range of
sources that can assess function and disability daily. 

Objective: This work aims to prove the feasibility of an interpretable machine learning pipeline for predicting function and
disability based on the World Health Organization Disability Assessment Schedule (WHODAS) 2.0 outcomes of clinical outpatients,
using passively collected digital biomarkers. 
Methods: One-month-long behavioral time-series data consisting of physical and digital activity descriptor variables were
summarized using statistical measures (minimum, maximum, mean, median, SD, and IQR), creating 64 features that were used
for prediction. We then applied a sequential feature selection to each WHODAS 2.0 domain (cognition, mobility, self-care, getting
along, life activities, and participation) in order to find the most descriptive features for each domain. Finally, we predicted the
WHODAS 2.0 functional domain scores using linear regression using the best feature subsets. We reported the mean absolute
errors and the mean absolute percentage errors over 4 folds as goodness-of-fit statistics to evaluate the model and allow for
between-domain performance comparison. 
Results: Our machine learning-based models for predicting patients' WHODAS functionality scores per domain achieved an
average (across the 6 domains) mean absolute percentage errors of 19.5%, varying between 14.86% (self-care domain) and 27.21% (life activities domain). We found that 5-19 features were sufficient for each domain, and the most relevant being the
distance traveled, time spent at home, time spent walking, exercise time, and vehicle time. 
Conclusions: Our findings show the feasibility of using machine learning-based methods to assess functional health solely
from passively sensed mobile data. The feature selection step provides a set of interpretable features for each domain, ensuring
better explainability to the models' decisions-an important aspect in clinical practice.
Published
2023
Pages
1–10
Journal
JMIR Formative Research, vol. 7, no. 2023, ISSN 2561-326X
DOI
10.2196/47167
UT WoS
001107546900002
EID Scopus
2-s2.0-85177448861
BibTeX 
@article{BUT186905,
  author="SUKEI, E. and ROMERO-MEDRANO, L. and DE LEON MARTINEZ, S. and HERRERA, J. and CAMPANA-MONTES, J. and OLMOS, M. and BACA-GARCIA, E. and ARTES, A.",
  title="Continuous Assessment of Function and Disability via Mobile Sensing: Real-World Data-Driven Feasibility Study",
  journal="JMIR Formative Research",
  year="2023",
  volume="7",
  number="2023",
  pages="1--10",
  doi="10.2196/47167",
  issn="2561-326X",
  url="https://formative.jmir.org/2023/1/e47167/authors"
}
Back to top