Chicago – Researchers at Northwestern University´s Fainberg School of Medicine published a new study which concludes that depression could be diagnosed, with almost 90% of accuracy, using algorithms to analyze mobile phone daily usage and GPS data.
Clinical psychologist, David Mohr and Computer Scientist, Sohrob Saeb, gathered information from the smartphones of 40 participants for two weeks. Using a sensor data acquisition app for Android, called Purple Robot, researchers were able to collect useful information from 28 participants only (20 females and 8 males, age ranging from 19 to 58). At the beginning of the experiment participants held a standardized self-reported depression survey called PHQ-9, which facilitates information about symptoms used to diagnose depression. Saeb stated that the phone data was more reliable in detecting depression than daily questions participants answered about how sad they were feeling on a scale.
“With phone sensor and GPS data, we can detect if a person has depressive symptoms and the severity of those symptoms without asking them any questions” said David Mohr, one of the authors.
In the paper, published on July 15 in the Journal of Medical Internet Research, authors reported that results obtained from the mobile phone sensor data analysis, showed correlation between daily usage (internet, apps, games), time spent in places like home or work, and symptoms of depression. Of the 28 participants, 50% did not have any signs of depression and 50% had symptoms ranging from mild to severe depression.
For instance, the study found that the average daily usage for depressed people was 68 minutes, while those without depression averaged 17 minutes. Moreover, patterns of location variance and home stay were used to measure whether the participants matched the normal symptoms of depression, given that people with this condition tend to move less through geographic space, and live a more sedentary live.
This part of the study showed stronger correlation. On this matter the research revealed, based on GPS data, that people who repeated their schedule day after day, had a good distribution of time in different places, and moved a lot in terms of physical distance were less likely to be depressed. Senior author David Mohr commented that “data showing depressed people tended not to go many places reflects the loss of motivation seen in depression“.
The conclusions of the publication, addressed the importance of replicating this study with a larger number of participants, given that technology and data analysis of this kind “offers numerous clinical opportunities, including continuous monitoring of at-risk populations with little patient burden and interventions that can provide just-in-time outreach“.
Leading author Sohrob Saeb expressed in an interview that, “If these methods are successful in finding out if someone has depression, symptoms won’t require any input from the patient. We’ll be able to passively and objectively measure behavior without a patient having to report this every day.“
In the future Saeb hopes to answer one question that appears to be crucial, “whether it is these behaviors that are causing the depression or whether the depression is causing the behaviors“.
Saeb remain ambiguous and says “it can be bidirectional“.
Source: Journal of Medical Internet Research