• Corpus ID: 221069791

Personalized Stress Detection with Self-supervised Learned Features

  title={Personalized Stress Detection with Self-supervised Learned Features},
  author={Stefan Matthes and Zhiwei Han and Tianming Qiu and Bruno Michel and S{\"o}ren Klinger and Hao Shen and Yuanting Liu and Bashar Altakrouri},
Automated stress detection using physiological sensors is challenging due to inaccurate labeling and individual bias in the sensor data. Previous methods consider stress detection as a supervised classification task, where bad labeling leads to a large performance drop. Furthermore, the poor generalizability to unseen subjects reveals the importance of personalizing stress detection for both interand intra-individual sensor data variability. Towards this end we present a label-free feature… 

Figures and Tables from this paper



Stress Detection Using Wearable Physiological and Sociometric Sensors

The experimental results show that by combining the measurements from both sensor systems, this paper could accurately discriminate between stressful and neutral situations during a controlled Trier social stress test (TSST).

Introducing WESAD, a Multimodal Dataset for Wearable Stress and Affect Detection

This work introduces WESAD, a new publicly available dataset for wearable stress and affect detection that bridges the gap between previous lab studies on stress and emotions, by containing three different affective states (neutral, stress, amusement).

Detecting Work Stress in Offices by Combining Unobtrusive Sensors

The focus of this paper is on developing automatic classifiers to infer working conditions and stress related mental states from a multimodal set of sensor data (computer logging, facial expressions, posture and physiology).

Subject-dependent biosignal features for increased accuracy in psychological stress detection

DeStress: Deep Learning for Unsupervised Identification of Mental Stress in Firefighters from Heart-rate Variability (HRV) Data

It is demonstrated that the clusters produced by the convolutional autoencoders consistently and successfully stratify stressed versus normal samples, as validated by several established physiological stress markers such as RMSSD, Max-HR, Mean-HR and LF-HF ratio.

HealthyOffice: Mood recognition at work using smartphones and wearable sensors

A novel mood recognition framework that is able to identify five intensity levels for eight different types of moods every two hours is proposed, and a smartphone app is presented, designed to facilitate self-reporting in a structured manner and provide the ground truth.

Wearable affect and stress recognition: A review

A broad overview and in-depth understanding of the theoretical background, methods, and best practices of wearable affect and stress recognition is provided, and psychological models, and detail affect-related physiological changes and their measurement with wearables are summarised.

Machine learning-based signal processing using physiological signals for stress detection

A signal processing approach is introduced based on machine learning algorithms for stress cognition and shows that the stress level can be detected by accuracy of 98.41% for 100 seconds and 200 seconds time intervals and 99% for 300 seconds time interval.

Stress Detection in Computer Users Based on Digital Signal Processing of Noninvasive Physiological Variables

  • Jing ZhaiA. Barreto
  • Computer Science
    2006 International Conference of the IEEE Engineering in Medicine and Biology Society
  • 2006
Results indicate that the physiological signals monitored do, in fact, have a strong correlation with the changes in emotional state of experimental subjects when stress stimuli are applied to the interaction environment.

The Influence of Person-Specific Biometrics in Improving Generic Stress Predictive Models

A simple, yet practical and cost effective calibration technique that derives an accurate and personalized stress prediction model from physiological samples collected from a large population and shows that it performs much better than a generic model.