Nivedit Majumdar Nivedit Majumdar

Quantified Self: Detecting and Resolving Depression by your mobile phone

An underlying purpose of technology is to better lives everywhere, be it in the form of ease of usage, information being available more readily or simply connecting with old friends. Technology is also prevalent in the field of healthcare and general medicine, with mHealth applications becoming more and more popular. Besides that, of course, is the whole Quantified Self movement of self tracking which involves self betterment through technology.

Apart from all these, however, is a more obscure and delicate aspect. The state of mental health is a crucial one, specially in this day and age of increasing workload and competition. Therefore, it is all the more vital for specialists to detect and resolve depression and other such mental disorders at the earliest, to improve the quality of life in general.

And this is the crux of my article here: How Quantified Self and smartphone applications can help in detecting whether the user has depression, and how these two sectors can aid in remedying the situation.

 

DEPRESSION – THE NUMBERS IN THIS REGARD

Depression and other mental health disorders such as anxiety and work related stress is quite a big problem. Sometimes, these cases are not just occasional, but they can fester to be lifelong problems which would hinder the concerned person from living a healthy life overall. Such mental disorders cause alarm for not just the concerned patient, but also to his/her near and dear ones.

In fact, the statistics in this regard are overwhelming. Here is a statistic showing the total expenditure in the U.S. for mental health services, from 2012 to 2014 and then a forecast from 2015 to 2020, and the rising graph is a huge cause of concern. It is estimated that in 2016, the  expenditure for mental health services will be to the tune of $194.4 million, which will further rise to about $238.4 million in 2020.

emberify_total_us_expenditure_for_mental_health_Services_2012_2020
(Data Source: Statista)

Besides this, lets look at the revenue of mental health practitioners’ offices in the U.S. In 2012, the revenue totalled to about $8 billion, which grew to about $9.9 billion in 2014, and is further expected to grow to about $10.6 billion in 2016 and $11.43 billion in 2020.

emberify_Revenue_of_mental_health_practitioners_offices_in_us_2012_2020
(Data Source: Statista)

Similar to mental health practitioners, are the mental health specialists. And here too, the numbers aren’t looking great for the overall state of depression in the U.S. From $4.8 billion in 2012, the revenue grew to about $5.41 billion in 2014, and is further expected to grow to about $5.6 billion in 2016 and $5.8 billion in 2020.

emberify_Revenue_of_mental_health_specialists_offices_in_us_2012_2020
(Data Source: Statista)

In the Great Britain too, the statistics aren’t good. In the period 2013/2014, 487 thousand workers suffered from work related stress, depression or anxiety. (Data Source: Statista) All in all, depression is a global problem. And pills and medical treatment need to be supplemented by some other factors to remedy the situation. Luckily, QS and smartphones prove to be the factors in this regard.

DETECTING DEPRESSION THROUGH SMARTPHONES

A new study, published in the Journal of Medical Internet Research in July 2015, states that it is quite possible to predict and detect depression through smartphone usage. The advantages of this method would be that the concerned person would be at his/her natural state, without getting conscious of something tracking him/her, which would in turn improve the efficiency of the findings.

The methods of correlating this research were actually ingenious, according to me. Daily life behavioural markers detected the GPS positioning of the target person, and usage sensors took readings. Factors such as phone usage features, usage duration and usage frequency were correlated and results were recorded.

emberify_detecting_depression_paper
(Image Source: Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study)

The results were quite brilliant. A number of features from GPS data were found to be related to depressive symptoms, including circadian movement, normalised entropy and location variance. Regression models and classifiers were utilised to improve the efficiency of the overall system to about 87%. The full research paper can be viewed here.

FACTORS RECORDED IN THE STUDY

To round up, three key factors were recorded, and on the basis of these one could come to conclusions regarding the mental health of the target person:

1. USAGE TIME:

Non depressed people, on an average, spent 17 minutes using their phone each day. Depressed people, on the other hand, consistently used their phones for over 68 minutes in a day.

Dr.Mohr, one of the authors of the study, said that this is an avoidance technique. People resort to using smartphones a lot when they try and avoid painful thoughts or other troubling emotions. Avoidance has been proved to be linked with depression.

2. CHANGES IN NORMAL SCHEDULES:

People’s regular routines were thrown out of balance when they faced depression, for example, they left for work at different times each day.

3. AVERSION TO TRAVEL:

This is where the GPS data comes in. People did not go to many places through the course of the study. According to Dr. Mohr, this reflects the loss of motivation and the tendency of people to withdraw and avoid going out and doing things – observations keenly linked with depression.

Personally, I think that these studies are relative to the user’s previous behaviours. If a person was outgoing, methodical and active previously, and displayed these symptoms after a certain point of time, and in a drastic manner at that, chances are that the person is going through some stress or anxiety. And these issues should be addressed before they fester into serious disorders such as depression.

THE LINK WITH INSTANT

Instant – is an application made by our team which tracks how much time you spend on your device. It also detects how much time you spend within applications and also generates graphs on usage history and correlates all this with location as well. You can study your behaviour using these statistics.

The relevance of this app is all the more prominent if one considers the factors I’ve talked about previously. Through applications such as Instant, one can visualise and easily understand the smartphone usage patterns. The tracking algorithms behind the application, aided with sensor technology within the smartphone (and the improvements too!) can aid detection of mental disorders and possible reduce the  chances of depression.

Sign up for our monthly mailing list