POLARstress: Lightweight Real-Time Stress Detection from Polar Verity Sense HR Data

About

POLARstress is a lightweight, HR-based stress detection algorithm designed for real-time continuous monitoring using Polar Verity Sense sensor data. It applies Heart Rate Variability (HRV) analysis to approximate stress levels using the RMSSD (Root Mean Square of Successive Differences) metric. The algorithm is optimized to work on low computational power devices, making it suitable for real-time applications in clinical settings such as therapy sessions and biofeedback interventions.

This project aims to detect stress, arousal, stability, and relaxation phases using only HR (bpm) signals, focusing on practical, deployable use in mobile apps or wearable devices.

---

🎯 Key Features

• Stress Detection Using HR Only: No need for complex PPG or ECG signals—works with simple HR data.
• Baseline Calculation: Automatic identification of the most relaxing 2-minute baseline period (if not provided).
• Overlapping Windows Analysis: Tracks RMSSD trends over time for continuous stress monitoring.
• Stress Flagging: Flags stress levels based on RMSSD thresholds (Stressed, Aroused, Stable, Relaxed).
• Efficient Computation: Designed to run on devices with limited processing power (e.g., wearables, mobile apps).

---

How It Works

The algorithm performs the following steps:

1. Data Processing: Processes Polar Verity Sense session files and enriches them with timestamps.
2. Outlier Removal & Interpolation: Removes RR interval outliers and fills gaps in the data.
3. Baseline Calculation: Identifies a baseline RMSSD period by selecting the most relaxing 2-minute window.
4. Overlapping Windows Analysis: Computes RMSSD values for overlapping time windows to monitor HRV changes.
5. Stress Flagging: Flags each period as High Stress, Moderate Stress, Normal, or Relaxed based on deviations from the baseline.

---

Repository Structure

POLARstress/
├── README.md
├── LICENSE.md
├── src/
│   ├── stress_detection_algo.py  # Main script for stress detection
├── data/
│   ├── sample_data1.csv      # Example Polar session data1
│   └── sample_data2.csv      # Example Polar session data1
└── requirements.txt

---

Script Details

stress_detection.py

The core script processes Polar Verity Sense session files to detect stress levels using HR-based RMSSD calculations. It performs:

• Data Enrichment
• Outlier Removal
• Baseline RMSSD Calculation
• Overlapping Window RMSSD Calculation
• Stress Flagging
• Visualization (Optional)

How Stress Levels Are Flagged

Stress levels are determined by comparing the calculated RMSSD values with the baseline RMSSD. The system flags the stress levels as follows:

Stress Level	Condition	Color in Plot
Stressed	RMSSD ≤ baseline_mean - 2 * baseline_std	Red
Aroused	baseline_mean - 2 * baseline_std < RMSSD ≤ baseline_mean - baseline_std	Yellow
Stable	baseline_mean - baseline_std < RMSSD ≤ baseline_mean + baseline_std	Green
Relaxed	RMSSD > baseline_mean + baseline_std	Blue

---

Usage:

python src/stress_detection.py --input_file <path/to/input.csv> --output_folder <path/to/output_folder> --save_plot

Sample data from session 1 and 2, output visualization:

---

Requirements

To install the required packages:

pip install -r requirements.txt

---

---

License

This project is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).

You are free to use, modify, and share the code for non-commercial purposes as long as proper attribution is given.

For any commercial use or inquiries, please contact the authors at [alvari.gianpaolo@gmail.com].

To view a copy of this license, visit: https://creativecommons.org/licenses/by-nc/4.0/

---

Usage of Polar Verity Sense Sensor Data

This project processes HR (Heart Rate) data obtained from the Polar Verity Sense sensor. Please note:

The software is not affiliated with or endorsed by Polar Electro.

Users are responsible for ensuring compliance with the Polar terms of use and privacy policy when using Polar devices and data.

For more information on Polar products, visit: https://www.polar.com

---

---

References

@article{alvari2024exploring,
  title={Exploring Physiological Responses in Virtual Reality-based Interventions for Autism Spectrum Disorder: A Data-Driven Investigation},
  author={Alvari, Gianpaolo and Vallefuoco, Ersilia and Cristofolini, Melanie and Salvadori, Elio and Dianti, Marco and Moltani, Alessia and Castello, Davide Dal and Venuti, Paola and Furlanello, Cesare},
  journal={arXiv preprint arXiv:2404.07159},
  year={2024}
}


@article{gabrielli2023co,
  title={Co-Design of a Virtual Reality Multiplayer Adventure Game for Adolescents With Autism Spectrum Disorder: Mixed Methods Study},
  author={Gabrielli, Silvia and Cristofolini, Melanie and Dianti, Marco and Alvari, Gianpaolo and Vallefuoco, Ersilia and Bentenuto, Arianna and Venuti, Paola and Ibarra, Oscar Mayora and Salvadori, Elio and others},
  journal={JMIR Serious Games},
  volume={11},
  number={1},
  pages={e51719},
  year={2023},
  publisher={JMIR Publications Inc., Toronto, Canada}
}