A multi-modal wearable dataset for cognitive attention and task-based stress analysis.
Mohammod Abdul Motin1, Md Santo Ali1, Sapnil Sarker Bipro1
1Department of Electrical & Electronic Engineering, Rajshahi University of Engineering & Technology, Rajshahi 6204, Bangladesh.
Data in Brief
|December 11, 2025
Summary
This study introduces the CATSA dataset for stress monitoring using wearable devices. It aids in developing AI for personalized stress management by analyzing physiological signals.
Area of Science:
- Physiology
- Biomedical Engineering
- Data Science
Background:
- Stress significantly impacts physical and mental health.
- Wearable devices enable real-time physiological monitoring for stress management.
- Personalized stress monitoring requires comprehensive physiological datasets.
Purpose of the Study:
- To introduce the CATSA multimodal dataset for stress monitoring.
- To facilitate the development of AI-based stress detection and management systems.
- To provide a resource for research on wearable sensor data analysis.
Main Methods:
- Collected multimodal physiological signals (blood volume pulse, electrodermal activity, acceleration, heart rate) from 50 participants.
- Utilized the Empatica E4 wrist-worn device for data acquisition.
- Employed a structured protocol to induce cognitive stress in participants.
Main Results:
- The CATSA dataset comprises diverse physiological signals under induced stress conditions.
- Data includes blood volume pulse, tri-axis acceleration, electrodermal activity, and average heart rate.
- The dataset is suitable for developing and validating stress monitoring algorithms.
Conclusions:
- The CATSA dataset is a valuable resource for advancing automated stress monitoring.
- It supports the creation of intelligent systems for personalized stress management.
- Facilitates research in signal processing and AI for physiological data analysis.
