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

Coefficients of correlation between location features. One, two, and three asterisks indicate significant correlation levels at P<.05, P<.01, and P<.001, respectively (ENT, entropy; ENTN, normalized entropy; LV, location variance; HS, home stay; TT, transition time; TD, total distance; CM, circadian movement; NC, number of clusters).
© Copyright Policy - open-access
Related In: Results  -  Collection

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

figure5: Coefficients of correlation between location features. One, two, and three asterisks indicate significant correlation levels at P<.05, P<.01, and P<.001, respectively (ENT, entropy; ENTN, normalized entropy; LV, location variance; HS, home stay; TT, transition time; TD, total distance; CM, circadian movement; NC, number of clusters).

Mentions: A correlation analysis across the features revealed that a number of them were highly correlated (Figure 5). Noticeably, there was a significant correlation between normalized entropy, location variance, and home stay. This is not surprising, as all these variables measure the amount of movement through space in different ways. However, the significant correlation between circadian movement and location variance is interesting and indicates that participants with more mobility also had more regular patterns of movement. The correlation between phone usage duration and frequency was also high (r=.89).


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)

Coefficients of correlation between location features. One, two, and three asterisks indicate significant correlation levels at P<.05, P<.01, and P<.001, respectively (ENT, entropy; ENTN, normalized entropy; LV, location variance; HS, home stay; TT, transition time; TD, total distance; CM, circadian movement; NC, number of clusters).
© Copyright Policy - open-access
Related In: Results  -  Collection

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

figure5: Coefficients of correlation between location features. One, two, and three asterisks indicate significant correlation levels at P<.05, P<.01, and P<.001, respectively (ENT, entropy; ENTN, normalized entropy; LV, location variance; HS, home stay; TT, transition time; TD, total distance; CM, circadian movement; NC, number of clusters).
Mentions: A correlation analysis across the features revealed that a number of them were highly correlated (Figure 5). Noticeably, there was a significant correlation between normalized entropy, location variance, and home stay. This is not surprising, as all these variables measure the amount of movement through space in different ways. However, the significant correlation between circadian movement and location variance is interesting and indicates that participants with more mobility also had more regular patterns of movement. The correlation between phone usage duration and frequency was also high (r=.89).

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