手机在一定程度上就是个体世界的缩影，有你的通讯录，日程表，定位，有你使用的apps。

现在发现手机使用记录都可以同来帮助判定使用者的精神状态，尤其是对于抑郁患者更显神功。

通过每天的使用记录和定位信息，可以在相当程度上判定个体是否存在抑郁症状：抑郁个体平均每天耗在手机上的时间远超过正常个体；更喜欢宅在家里或是活动场所相对固定者更容易呈现抑郁症状。

原文链接：http://www.jmir.org/2015/7/e175/

文体：Mobile Phone Sensor Correlates of Depressive Symptom Severity inDaily-Life Behavior: An Exploratory Study

ABSTRACT

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

Objective: The objective of this study wasto explore the detection of daily-life behavioral markers using mobile phoneglobal positioning systems (GPS) and usage sensors, and their use inidentifying depressive symptom severity.

Methods: A total of 40 adult participantswere recruited from the general community to carry a mobile phone with a sensordata acquisition app (Purple Robot) for 2 weeks. Of these participants, 28 hadsufficient sensor data received to conduct analysis. At the beginning of the2-week period, participants completed a self-reported depression survey(PHQ-9). Behavioral features were developed and extracted from GPS location andphone usage data.

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

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

群晓生物

qbioscience.com