JMIR Publications

ABSTRACT

Background:

Healthcare, in recent years, has made great leaps in integrating wireless technology into traditional models of care. The availability of ubiquitous devices such as wearable sensors has enabled researchers to collect voluminous datasets and to harness them in a wide range of healthcare topics. One of the goals of using on-body wearable sensors has been to study and analyze human activity and functional patterns, thereby predicting harmful outcomes such as falls. It can also be used to track precise individual movements to form personalized behavioral patterns, standardize the concept of frailty, well-being/independence and many other applications. Most wearable devices such as activity trackers and smartwatches are equipped with low cost embedded sensors (e.g., accelerometer, gyroscope, barometer, heart rate monitor) that can provide users with health statistics. In addition to wearable devices, Bluetooth low energy sensors known as BLE beacons have gained traction among researchers in ambient intelligence domain. The low cost and durability of newer versions have made BLE beacons feasible gadgets to yield indoor localization data; an adjunct feature in human activity recognition. In [1]-[4], we investigated and introduced a generic framework (Sensing At-Risk Population - SARP) that draws upon classification of human movements using a 3-axial accelerometer and extracting indoor localization using BLE beacons, in concert.

Objective:

To examine combination of physical activities and indoor locations of patients extracted at baseline on a cohort of 154 rehabilitation-dwelling patients. We aimed to explore the capability of significant physical activity and indoor localization features to discriminate between subacute care patients that are readmitted to the hospital versus the patients who are able to stay in a community setting.

Methods:

We analyzed physical activity sensor features to assess activity time and intensity. We also analyzed activities with respect to indoor localization. χ^2and Kruskal-Wallis tests were used to compare demographic variables and sensor feature variables in outcome groups. Random forests were used to build predictive models based on the most significant features.

Results:

Standing time percentage (P < .001, d=1.51), laying down time percentage (P < .001, d=1.35), resident room energy intensity (P < .001, d=1.25), resident bed energy intensity (P < .001, d=1.23) and energy percentage of active state (P = .001, d=1.24) are the five most statistically significant features in distinguishing outcome groups at baseline. Energy intensity of resident room (P < .001, d=1.25) was achieved by capturing indoor localization information. Random forests revealed that energy intensity of resident room, as a standalone attribute, is the most sensitive parameters in identification of outcome groups (AUC=0.84).

Conclusions:

This study demonstrates that combination of indoor localization and physical activity tracking produces a series of features at baseline, a subset of which can better distinguish between at-risk patients that can gain independence versus the patients that are re-hospitalized.