< Back to previous page

Project

Smart algorithms for wearables to reduce relapse into stress related disorders

To date, depression is the worldwide leading cause for disability and a considerable contributor to the global disease burden. In recent years, another disorder that shares certain traits with depression has been on the rise, namely burnout. Although the prevalence and disease burden of burnout is not nearly as great as that of depression, it has been escalating in the last decade. As a result, burnout has been receiving a lot of attention in the media and companies have started putting their resources to detect and prevent burnout in the workplace. Both disorders currently have a major impact on Western society and cannot be thought away and could thus be called ‘21st century epidemics’.

Whereas available literature on depression is vast, literature on burnout is scarce and literature comparing depression to burnout is even more limited. There is still a lot to learn about burnout and how it compares to depression. This is especially true when it comes to studying the psychophysiology of both disorders. For instance, the search for biomarkers to quantify depression has been going on for several decades, whereas it has only just started for burnout. Although we haven’t found a ‘true biomarker’ for depression yet, we already have a great knowledge of its physiology. This knowledge is lacking for burnout.

The general aim of this PhD thesis was to study the psychophysiology of patients with burnout or depression and to compare both patient groups, as well as to compare them to healthy controls. For that purpose, a clinical study was set up where healthy controls, patients with burnout and patients with depression performed a combined physical and mental task. All participants cycled for half an hour and completed an arithmetic challenge of three minutes during cycling. The energy expenditure of the participants was measured via indirect calorimetry. Their heart rate was monitored with an ECG and with a wearable sports watch. Finally, the delivered biking power and the participants’ core body acceleration were logged as well.

Using data-based mechanistic input-output modelling with power as input and energy expenditure or heart rate as output, it was demonstrated that the energy expenditure and heart rate of healthy controls and patients with burnout or depression can be equivalently decomposed into a physical component and a mental component. A detailed study of the energy expenditure revealed that the energy metabolism of patients with burnout or depression is dysregulated and that this dysregulation seems to be more important for depressive patients. A thorough evaluation of the heart rate response showed that the link between power and heart rate is altered in patients with burnout or depression. Interestingly, we found that features derived from the mental component of energy expenditure or heart rate could not differentiate between the three subject groups, whereas features derived from the physical component could. Finally the data-based mechanistic input-output modelling approach was also successfully applied with activity (derived from acceleration) as input and heart rate from a sports watch as output. We found that the heart rate data of the sports watch were as accurate as the heart rate data of ECG. Additionally, we observed the same trends in the features obtained from the wearable data as the trends found in the features calculated from the reference data.

Summarised, this PhD thesis first and foremost demonstrates the universality of the data-based mechanistic input-output modelling approach. Second, this work shows that during physical exercise the energy metabolism and the heart rate response of patients with burnout or depression is dysregulated. Third, the results of this research suggest that more research should be conducted on the physiological response of burnout patients and depressive patients to a physical exercise, rather than a mental exercise. Finally, this PhD illustrates that it is feasible to monitor the physiology of burnout and depression using wearable technology. 

Date:10 Aug 2015 →  6 May 2020
Keywords:Smart algorithms, wearables, stress related disorders
Disciplines:Psychiatry and psychotherapy, Nursing, Other paramedical sciences, Clinical and counselling psychology, Other psychology and cognitive sciences
Project type:PhD project