A 2012 survey conducted by the World Health Organization (WHO) suggests that depression is a major public health issue that affects 350 million individuals globally. Depression when undiagnosed and untreated, is detrimental to physical health, social relations and quality of life and can also lead to self harm and suicide. Depression is a leading cause of disability around the world and contributes greatly to the global burden of disease. The effects of depression can be long-lasting or recurrent and can dramatically affect a person's ability to function and live a rewarding life. It is associated with disrupted biological rhythms caused by environmental disturbance like seasonal change in daylight, alteration of social rhythms due to for instance shift-work or longitude traveling; besides linked to lifestyles associated with diurnal rhythms inconsistent with the natural daylight cycle. Before depression can be effectively treated and it has to be rightfully recognised. Recognising depression in patients is an ongoing battle due to the complexity of several features used to correctly classify depression. Moreover, the research on identifying depression through motion sensing data is relatively new. In this project we seek to develop various classifiers on a dataset of user information collected through a general survey and motor activity data collected through a smartwatch to find the best algorithm that results in the most accurate classification. Our results will be really valuable in addressing key concerns in countering depression.
For the purpose of this project, we would rely on the following definition of Depression: A mental health disorder characterized by persistent feelings of sadness or loss of interest in activities, causing significant impairment in daily life. Possible causes include a combination of biological, psychological and social sources of distress.
- Muskaan Kumar
- Kushal Mishra
- Rya Sanovar
- Shreya Senapaty