A Novel Approach to Clustering Accelerometer Data for Application in Passive Predictions of Changes in Depression Severity.

Autor: Ross MK; Department of Psychiatry, University of Illinois at Chicago, Chicago, IL 60612, USA.; Department of Biomedical Engineering, University of Illinois at Chicago, Chicago, IL 60612, USA., Tulabandhula T; Department of Information and Decision Sciences, University of Illinois at Chicago, Chicago, IL 60612, USA., Bennett CC; Department of Intelligence Computing, Hanyang University, Seoul 04763, Republic of Korea.; Department of Computing, DePaul University, Chicago, IL 60604, USA., Baek E; Department of Intelligence Computing, Hanyang University, Seoul 04763, Republic of Korea., Kim D; Department of Intelligence Computing, Hanyang University, Seoul 04763, Republic of Korea., Hussain F; Department of Psychiatry, University of Illinois at Chicago, Chicago, IL 60612, USA., Demos AP; Department of Psychology, University of Illinois at Chicago, Chicago, IL 60612, USA., Ning E; Department of Psychology, University of Illinois at Chicago, Chicago, IL 60612, USA., Langenecker SA; Department of Psychiatry, University of Utah, Salt Lake City, UT 84112, USA., Ajilore O; Department of Psychiatry, University of Illinois at Chicago, Chicago, IL 60612, USA., Leow AD; Department of Psychiatry, University of Illinois at Chicago, Chicago, IL 60612, USA.; Department of Biomedical Engineering, University of Illinois at Chicago, Chicago, IL 60612, USA.
Jazyk: angličtina
Zdroj: Sensors (Basel, Switzerland) [Sensors (Basel)] 2023 Feb 01; Vol. 23 (3). Date of Electronic Publication: 2023 Feb 01.
DOI: 10.3390/s23031585
Abstrakt: The treatment of mood disorders, which can become a lifelong process, varies widely in efficacy between individuals. Most options to monitor mood rely on subjective self-reports and clinical visits, which can be burdensome and may not portray an accurate representation of what the individual is experiencing. A passive method to monitor mood could be a useful tool for those with these disorders. Some previously proposed models utilized sensors from smartphones and wearables, such as the accelerometer. This study examined a novel approach of processing accelerometer data collected from smartphones only while participants of the open-science branch of the BiAffect study were typing. The data were modeled by von Mises-Fisher distributions and weighted networks to identify clusters relating to different typing positions unique for each participant. Longitudinal features were derived from the clustered data and used in machine learning models to predict clinically relevant changes in depression from clinical and typing measures. Model accuracy was approximately 95%, with 97% area under the ROC curve (AUC). The accelerometer features outperformed the vast majority of clinical and typing features, which suggested that this new approach to analyzing accelerometer data could contribute towards unobtrusive detection of changes in depression severity without the need for clinical input.
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje