Comparison of Machine Learning Techniques for Psychophysiological Stress Detection

Smets, Elena; Casale, Pierluigi; Großekathöfer, Ulf; Lamichhane, Bishal; Raedt, W. De; Bogaerts, Katleen; Diest, Ilse Van; Hoof, Chris Van

doi:10.1007/978-3-319-32270-4_2

Cited by 53 publications

(43 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Hospital for Sick Children studied different big data analytic methods to process several vital signs of patients in order to predict hospital infection up to 24 hours in advance than traditional methods. In addition, several machine learning techniques have been proposed automatically to detect psycho-physiological stress from bio-signals such as accelerometer and skin conductance [121], [122]. The authors of [118], constructed a behavior model based on time series data using machine learning techniques for fast detection of abnormality.…”

Section: H Population Health Managementmentioning

confidence: 99%

Towards fog-driven IoT eHealth: Promises and challenges of IoT in medicine and healthcare

Farahani

Firouzi

Chang

et al. 2018

Future Generation Computer Systems

815

371

View full text Add to dashboard Cite

Internet of Things (IoT) offers a seamless platform to connect people and objects to one another for enriching and making our lives easier. This vision carries us from compute-based centralized schemes to a more distributed environment offering a vast amount of applications such as smart wearables, smart home, smart mobility, and smart cities. In this paper we discuss applicability of IoT in healthcare and medicine by presenting a holistic architecture of IoT eHealth ecosystem. Healthcare is becoming increasingly difficult to manage due to insufficient and less effective healthcare services to meet the increasing demands of rising aging population with chronic diseases. We propose that this requires a transition from the clinic-centric treatment to patient-centric healthcare where each agent such as hospital, patient, and services are seamlessly connected to each other. This patient-centric IoT eHealth ecosystem needs a multi-layer architecture: 1) device, 2) fog computing and 3) cloud to empower handling of complex data in terms of its variety, speed, and latency. This fog-driven IoT architecture is followed by various case examples of services and applications that are implemented on those layers. Those examples range from mobile health, assisted living, e-medicine, implants, early warning systems, to population monitoring in smart cities. We then finally address the challenges of IoT eHealth such as data management, scalability, regulations, interoperability, device-network-human interfaces, security, and privacy.

show abstract

Section: H Population Health Managementmentioning

confidence: 99%

Towards fog-driven IoT eHealth: Promises and challenges of IoT in medicine and healthcare

Farahani

Firouzi

Chang

et al. 2018

Future Generation Computer Systems

815

371

View full text Add to dashboard Cite

show abstract

“…From an engineering point of view, this kind of approach is often used to extract features to train classifiers for the detection of particular physiological/mental states. Examples are studies concerning emotional recognition [ 10 , 11 ] or stress detection [ 12 , 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Information Dynamics of the Brain, Cardiovascular and Respiratory Network during Different Levels of Mental Stress

Zanetti

Faes

Nollo

et al. 2019

Entropy

View full text Add to dashboard Cite

In this study, an analysis of brain, cardiovascular and respiratory dynamics was conducted combining information-theoretic measures with the Network Physiology paradigm during different levels of mental stress. Starting from low invasive recordings of electroencephalographic, electrocardiographic, respiratory, and blood volume pulse signals, the dynamical activity of seven physiological systems was probed with one-second time resolution measuring the time series of the δ , θ , α and β brain wave amplitudes, the cardiac period (RR interval), the respiratory amplitude, and the duration of blood pressure wave propagation (pulse arrival time, PAT). Synchronous 5-min windows of these time series, obtained from 18 subjects during resting wakefulness (REST), mental stress induced by mental arithmetic (MA) and sustained attention induced by serious game (SG), were taken to describe the dynamics of the nodes composing the observed physiological network. Network activity and connectivity were then assessed in the framework of information dynamics computing the new information generated by each node, the information dynamically stored in it, and the information transferred to it from the other network nodes. Moreover, the network topology was investigated using directed measures of conditional information transfer and assessing their statistical significance. We found that all network nodes dynamically produce and store significant amounts of information, with the new information being prevalent in the brain systems and the information storage being prevalent in the peripheral systems. The transition from REST to MA was associated with an increase of the new information produced by the respiratory signal time series (RESP), and that from MA to SG with a decrease of the new information produced by PAT. Each network node received a significant amount of information from the other nodes, with the highest amount transferred to RR and the lowest transferred to δ , θ , α and β . The topology of the physiological network underlying such information transfer was node- and state-dependent, with the peripheral subnetwork showing interactions from RR to PAT and between RESP and RR, PAT consistently across states, the brain subnetwork resulting more connected during MA, and the subnetwork of brain–peripheral interactions involving different brain rhythms in the three states and resulting primarily activated during MA. These results have both physiological relevance as regards the interpretation of central and autonomic effects on cardiovascular and respiratory variability, and practical relevance as regards the identification of features useful for the automatic distinction of different mental states.

show abstract

“…However, by combining the response features of HR, GSR, and ST, the performance can be increased, and insights from all physiological signals can be obtained. Furthermore, previous research has suggested that the physiological stress response is person‐dependent, and different participants can show different response levels per physiological signal . For these reasons, and since HR, GSR, and ST are standard measurements readily available in many state‐of‐the art sensors, eg, NeXus 10 MK II, and in multiple wearables such as Empatica E4 (Empatica, Milan, Italy), it is advised to focus further research on the combination of these signals rather than investigate them separately.…”

Section: Discussionmentioning

confidence: 99%

“…For the healthy participants, the protocol additionally included a counting task before the first rest phase and after the last rest phase, as presented in detail by Smets et al The goal of this counting task was to control for the physiological response to speaking. We have shown that a stressful task with speech can be distinguished from a nonstressful speaking task, ie, counting . Since the counting task did not significantly differ from a rest phase, it was removed to reduce the experimental time for the patients.…”

Section: Methodsmentioning

confidence: 99%

Comparing task‐induced psychophysiological responses between persons with stress‐related complaints and healthy controls: A methodological pilot study

Smets

Schiavone

Velazquez

et al. 2018

Health Science Reports

Self Cite

View full text Add to dashboard Cite

AimsChronic stress is an important factor for a variety of health problems, highlighting the importance of early detection of stress‐related problems. This methodological pilot study investigated whether the physiological response to and recovery from a stress task can differentiate healthy participants and persons with stress‐related complaints.Methods and ResultsHealthy participants (n = 20) and participants with stress‐related complaints (n = 12) participated in a laboratory stress test, which included 3 stress tasks. Three physiological signals were recorded: galvanic skin response (GSR), heart rate (HR), and skin temperature (ST). From these signals, 126 features were extracted, including static (eg, mean) and dynamic (eg, recovery time) features. Unsupervised feature selection reduced the set to 26 features. A logistic regression model was developed for 6 feature sets, analysing single‐parameter and multiparameter models as well as models using recovery vs response‐related features. The highest classification performance (accuracy = 78%) was obtained using the response‐related feature set, including all physiological signals and using GSR‐related features. A worse performance was obtained using single‐signal feature sets based on HR (accuracy = 66%) and ST (accuracy = 59%). Response‐related features outperformed recovery‐related features (accuracy = 63%).ConclusionParticipants with stress‐related complaints may be differentiated from healthy controls by physiological responses to stress tasks. We aimed to bring attention to new exploratory methodologies; further research is needed to validate and replicate the results on larger populations and patients on different areas along the stress continuum.

show abstract

Comparison of Machine Learning Techniques for Psychophysiological Stress Detection

Cited by 53 publications

References 21 publications

Towards fog-driven IoT eHealth: Promises and challenges of IoT in medicine and healthcare

Towards fog-driven IoT eHealth: Promises and challenges of IoT in medicine and healthcare

Information Dynamics of the Brain, Cardiovascular and Respiratory Network during Different Levels of Mental Stress

Comparing task‐induced psychophysiological responses between persons with stress‐related complaints and healthy controls: A methodological pilot study

Contact Info

Product

Resources

About