Limits...
Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study.

Saeb S, Zhang M, Karr CJ, Schueller SM, Corden ME, Kording KP, Mohr DC - J. Med. Internet Res. (2015)

Bottom Line: Phone usage features, usage duration, and usage frequency were also correlated (r=.54, P=.011, and r=.52, P=.015, respectively).Furthermore, a regression model that used the same feature to estimate the participants' PHQ-9 scores obtained an average error of 23.5%.While these findings must be replicated in a larger study among participants with confirmed clinical symptoms, they suggest that phone sensors offer numerous clinical opportunities, including continuous monitoring of at-risk populations with little patient burden and interventions that can provide just-in-time outreach.

View Article: PubMed Central - HTML - PubMed

Affiliation: Center for Behavioral Intervention Technologies, Department of Preventive Medicine, Northwestern University, Chicago, IL, United States.

ABSTRACT

Background: Depression is a common, burdensome, often recurring mental health disorder that frequently goes undetected and untreated. Mobile phones are ubiquitous and have an increasingly large complement of sensors that can potentially be useful in monitoring behavioral patterns that might be indicative of depressive symptoms.

Objective: The objective of this study was to explore the detection of daily-life behavioral markers using mobile phone global positioning systems (GPS) and usage sensors, and their use in identifying depressive symptom severity.

Methods: A total of 40 adult participants were recruited from the general community to carry a mobile phone with a sensor data acquisition app (Purple Robot) for 2 weeks. Of these participants, 28 had sufficient sensor data received to conduct analysis. At the beginning of the 2-week period, participants completed a self-reported depression survey (PHQ-9). Behavioral features were developed and extracted from GPS location and phone usage data.

Results: A number of features from GPS data were related to depressive symptom severity, including circadian movement (regularity in 24-hour rhythm; r=-.63, P=.005), normalized entropy (mobility between favorite locations; r=-.58, P=.012), and location variance (GPS mobility independent of location; r=-.58, P=.012). Phone usage features, usage duration, and usage frequency were also correlated (r=.54, P=.011, and r=.52, P=.015, respectively). Using the normalized entropy feature and a classifier that distinguished participants with depressive symptoms (PHQ-9 score ≥5) from those without (PHQ-9 score <5), we achieved an accuracy of 86.5%. Furthermore, a regression model that used the same feature to estimate the participants' PHQ-9 scores obtained an average error of 23.5%.

Conclusions: Features extracted from mobile phone sensor data, including GPS and phone usage, provided behavioral markers that were strongly related to depressive symptom severity. While these findings must be replicated in a larger study among participants with confirmed clinical symptoms, they suggest that phone sensors offer numerous clinical opportunities, including continuous monitoring of at-risk populations with little patient burden and interventions that can provide just-in-time outreach.

No MeSH data available.


Related in: MedlinePlus

Example GPS location data, overlaid on satellite image. Each small circle represents a histogram bin, which has a size of 500 by 500 meters. The colors indicate the number of samples captured by each bin (brighter means more samples). The bigger blue circles show the center of the clusters detected by the clustering algorithm.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4526997&req=5

figure1: Example GPS location data, overlaid on satellite image. Each small circle represents a histogram bin, which has a size of 500 by 500 meters. The colors indicate the number of samples captured by each bin (brighter means more samples). The bigger blue circles show the center of the clusters detected by the clustering algorithm.

Mentions: For the location data (Figure 1), we used two procedures. The first procedure determined whether each GPS location data sample came from a stationary state (eg, working in an office) or a transition state (eg, walking on the street). To do so, we estimated the movement speed at each location data sample by calculating its time derivative and then used a threshold speed that defined the boundary between these two states. In this study, we set this threshold to 1 km/h.


Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study.

Saeb S, Zhang M, Karr CJ, Schueller SM, Corden ME, Kording KP, Mohr DC - J. Med. Internet Res. (2015)

Example GPS location data, overlaid on satellite image. Each small circle represents a histogram bin, which has a size of 500 by 500 meters. The colors indicate the number of samples captured by each bin (brighter means more samples). The bigger blue circles show the center of the clusters detected by the clustering algorithm.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4526997&req=5

figure1: Example GPS location data, overlaid on satellite image. Each small circle represents a histogram bin, which has a size of 500 by 500 meters. The colors indicate the number of samples captured by each bin (brighter means more samples). The bigger blue circles show the center of the clusters detected by the clustering algorithm.
Mentions: For the location data (Figure 1), we used two procedures. The first procedure determined whether each GPS location data sample came from a stationary state (eg, working in an office) or a transition state (eg, walking on the street). To do so, we estimated the movement speed at each location data sample by calculating its time derivative and then used a threshold speed that defined the boundary between these two states. In this study, we set this threshold to 1 km/h.

Bottom Line: Phone usage features, usage duration, and usage frequency were also correlated (r=.54, P=.011, and r=.52, P=.015, respectively).Furthermore, a regression model that used the same feature to estimate the participants' PHQ-9 scores obtained an average error of 23.5%.While these findings must be replicated in a larger study among participants with confirmed clinical symptoms, they suggest that phone sensors offer numerous clinical opportunities, including continuous monitoring of at-risk populations with little patient burden and interventions that can provide just-in-time outreach.

View Article: PubMed Central - HTML - PubMed

Affiliation: Center for Behavioral Intervention Technologies, Department of Preventive Medicine, Northwestern University, Chicago, IL, United States.

ABSTRACT

Background: Depression is a common, burdensome, often recurring mental health disorder that frequently goes undetected and untreated. Mobile phones are ubiquitous and have an increasingly large complement of sensors that can potentially be useful in monitoring behavioral patterns that might be indicative of depressive symptoms.

Objective: The objective of this study was to explore the detection of daily-life behavioral markers using mobile phone global positioning systems (GPS) and usage sensors, and their use in identifying depressive symptom severity.

Methods: A total of 40 adult participants were recruited from the general community to carry a mobile phone with a sensor data acquisition app (Purple Robot) for 2 weeks. Of these participants, 28 had sufficient sensor data received to conduct analysis. At the beginning of the 2-week period, participants completed a self-reported depression survey (PHQ-9). Behavioral features were developed and extracted from GPS location and phone usage data.

Results: A number of features from GPS data were related to depressive symptom severity, including circadian movement (regularity in 24-hour rhythm; r=-.63, P=.005), normalized entropy (mobility between favorite locations; r=-.58, P=.012), and location variance (GPS mobility independent of location; r=-.58, P=.012). Phone usage features, usage duration, and usage frequency were also correlated (r=.54, P=.011, and r=.52, P=.015, respectively). Using the normalized entropy feature and a classifier that distinguished participants with depressive symptoms (PHQ-9 score ≥5) from those without (PHQ-9 score <5), we achieved an accuracy of 86.5%. Furthermore, a regression model that used the same feature to estimate the participants' PHQ-9 scores obtained an average error of 23.5%.

Conclusions: Features extracted from mobile phone sensor data, including GPS and phone usage, provided behavioral markers that were strongly related to depressive symptom severity. While these findings must be replicated in a larger study among participants with confirmed clinical symptoms, they suggest that phone sensors offer numerous clinical opportunities, including continuous monitoring of at-risk populations with little patient burden and interventions that can provide just-in-time outreach.

No MeSH data available.


Related in: MedlinePlus