What question did this study set out to answer?

Introduce and evaluate WESAD, a publicly available multimodal dataset recorded from chest- and wrist-worn sensors to benchmark wearable stress and affect detection across multiple emotional states.

October 2, 2018

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

View Full Paper

Ask AI

Bookmark

Introduce and evaluate WESAD, a publicly available multimodal dataset recorded from chest- and wrist-worn sensors to benchmark wearable stress and affect detection across multiple emotional states.
Recorded physiological signals (ECG, BVP, EDA, EMG, respiration, temperature) and 3-axis acceleration from wrist and chest devices in N=15 participants during a laboratory protocol.
Elicited three affective states (neutral baseline, stress, and amusement) and collected validated psychological self-reports alongside continuous sensor streams.
Benchmarked standard machine learning classifiers for binary (stress vs. non-stress) and 3-class (neutral vs. stress vs. amusement) recognition, comparing chest and wrist modalities.
Standard machine learning models achieved classification accuracies of up to 80% for the three-class affect classification task (baseline vs. stress vs. amusement).
Binary classification distinguishing stress from non-stress conditions reached classification accuracies of up to 93% using multimodal sensor features.

Structured PICO

Population

15 subjects during a lab study

Intervention

Wearable stress and affect detection using wrist- and chest-worn devices recording physiological and motion data (blood volume pulse, electrocardiogram, electrodermal activity, electromyogram, respiration, body temperature, and three-axis acceleration)

Outcome

Classification accuracy for affective states (neutral, stress, amusement)surrogate

The WESAD dataset provides a multimodal benchmark for wearable stress and affect detection, achieving up to 93% accuracy for binary stress classification.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Wearable stress classifiers require clinical caution before adoption; leaves open prospective validation in cardiovascular cohorts.

Cite This Study

Schmidt et al. (2018) studied this question.

synapsesocial.com/papers/69d56e2175589c71d767d416 https://doi.org/10.1145/3242969.3242985

View Full Paper

Ask AI

Bookmark

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

Key Points

Structured PICO

Discussion

Implication

Cite This Study

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

Key Points

Structured PICO

Discussion

Implication

Cite This Study

Also Consider

Also Consider