The onset of stress triggers sympathetic arousal (SA), which causes detectable changes to physiological parameters such as heart rate, blood pressure, dilation of the pupils and sweat release. The objective quantification of SA has tremendous potential to prevent and manage psychological disorders. Photoplethysmography (PPG), a non-invasive method to measure skin blood flow changes, has been used to estimate SA indirectly. However, the impact of various wavelengths of the PPG signal has not been investigated for estimating SA. In this study, we explore the feasibility of using various statistical and nonlinear features derived from peak-to-peak (AC) values of PPG signals of different wavelengths (green, blue, infrared and red) to estimate stress-induced changes in SA and compare their performances. The impact of two physical stressors, Cold Pressor and Hand Grip, is studied on 32 healthy individuals. The results show that the nonlinear features are the most promising in detecting stress-induced sympathetic activity. TotalSampEn feature was capable of detecting stress-induced changes in SA for all wavelengths, whereas other features (Petrosian, AvgSampEn) are significant (AUC≥0.8$) only for IR and Red wavelengths. The outcomes of this study can be used to make device design decisions as well as develop stress detection algorithms.

The data consists of the computed nonlinear, linear and fractal statistical features from all subjects. These are saved as CSV files for two cohorts, SSNA and MSNA. 

The data is sorted into MSNA and SSNA as it is being reused for a different publication. But for the reader's info, the acronyms stand for muscle (MSNA) and skin sympathetic nerve activity (SSNA), respectively. For the analysis, these should be combined, as the separation of the features is not relevant to this investigation.

The code consists of loading data from the CSV files and computing the p-values that correspond to Table 4.

Software: 

Users should install the Python environment. The “requirements.txt” file includes all required python libraries.

Download the dataset and you can find the following zip file

• DATA_AND_CODE.zip
  •  data
    • MSNA_app_entropy.csv
    • SSNA_app_entropy.csv
  •   data_processing
    •   data_process.py
  • p_value_dir
    • AUC_value.csv
    • P_value_significance.csv
  •  error_bar_dir
    • HG.png
    • CP.png
  • main.py
  • requirements.txt
  • Read_me_file.pdf

File descriptions:

• data folder includes all the datasets and must be within your working directory for the analysis to work.
  •  Example: MSNA_app_entropy.csv = ”MSNA (muscle skin sympathetic nerve activity)” _” feature name” “.csv “extension 
• data_process.py: this file includes p-value and errorbar generator functions with proper documentation.
• AUC_value.csv: this file includes the significance of the AUC (area under the curve) value for different events such as a non-stress event to stress event (AC) (Table 5 in the manuscript shows the same value as mentioned in this file). 
• P_value_significance.csv: this file includes the significance of p_value for different events such as pre-stress to cold pressor to recovery (ABC) (Table 4 in the manuscript shows the same value as mentioned in this file).
• Error_bar_dir.zip: this file includes an errorbar for different features such as Mean, SD, Katz, Petrosian, Higuchi, app_entropy, sample_entropy, TotalSampEn, AvgSampEn (Figure 5 to 7 in the manuscript shows the same value as mentioned in this file).
• Main.py generates the result for AUC, p_value and error_bar into the p_value_dir and error_bar_dir folder.
• Requirement.txt: all required python libraries need to be installed.
• Read_me_file.pdf includes all directions for users.

 Note: Please cite this paper and collaborate with the authors if you are using this dataset.