Skip to main content
Dryad logo

A multi-modal sensor dataset for continuous stress detection of nurses in a hospital

Citation

Hosseini, Seyedmajid et al. (2021), A multi-modal sensor dataset for continuous stress detection of nurses in a hospital, Dryad, Dataset, https://doi.org/10.5061/dryad.5hqbzkh6f

Abstract

Advances in wearable technologies provide the opportunity to monitor many physiological variables continuously. Stress detection has gained increased attention in recent years, especially because early stress detection can help individuals better manage health to minimize the negative impacts of long-term stress exposure. This paper provides a unique stress detection dataset created in a natural working environment in a hospital. This dataset is a collection of biometric data of nurses during the COVID-19 outbreak. Studying stress in a work environment is complex due to the influence of many social, cultural, and individuals experience in dealing with stressful conditions. In order to address these concerns, we captured both the physiological data and associated context pertaining to the stress events. We monitored specific physiological variables, including electrodermal activity, heart rate, skin temperature, and accelerometer data of the nurse subjects. A periodic smartphone-administered survey also captured the contributing factors for the detected stress events. A database containing the signals, stress events, and survey responses is available upon request.

Methods

The data was gathered for approximately one week from 15 female nurses working regular shifts at a hospital. 1,250 hours worth of data was collected in two study sessions in Apr-May and Nov-Dec of 2020. The data was collected using Empatica E4 wearable devices. A survey was administered every day to identify the type of stress.

Usage Notes

Data description

The following is a description of directories and files in the dataset.

Stress_dataset.zip: The zip file holds the data of 15 participants in different folders. Each folder contains raw data signals in CSV format in a sub-folder. A raw data folder consists of 6 different CSV files, including (1) EDA.csv (electrodermal activity), (2) HR.csv (heart rate), (3) TEMP.csv (skin temperature), (4) IBI.csv (inter-beat interval), (5) BVP.csv (blood volume pulse), and (6) ACC.csv (accelerometer data).

Each biometric signal data has the following information:

  • Start time (epoch): The DateTime float number that contains the time that signal was generated using the internal clock of the wristband. The DateTime is stored at the first row of every data column.
  • Frequency: The second cell of each column shows the data collection frequency

ACC.csv:

  • Column I: x-Axis acceleration
  • Column II: y-Axis acceleration
  • Column III: z-Axis acceleration

BVP.csv:

  • Column I: Blood volume pulse is a method of measuring the heart rate.

EDA.csv:

  • Column I: Electrodermal activity of the skin, measuring the skin’s electrical conductivity.

IBI.csv:

  • Column I: time interval
  • Column II: Inter-beat interval or beat-to-beat interval, being the time interval between individual beats of an individual’s heart.

TEMP.csv:

  • Column I: Skin temperature in Celsius.

tags.csv: contains the timestamp of the user tag. A tag event occurs when the user clicks the button on the watch to mark an event. However, the subjects did not consistently use this feature and the field has no information value in our study.

In some cases, the sensor data csv files are empty. This constitutes a failure of the device to capture data.

Survey Results File

Each folder name is identical to the participant’s ID in both data and survey files. All of the signals were synchronized to bring them to a common frequency. The accelerometer data is not used in the stress detection model. Some of the basic physical activities can be estimated from the accelerometer sensor, which could be further used to potentially include the activity context in stress detection.

SurveyResults.xlsx: The Excel file holds all participant survey results and their annotated stress level in Excel sheets (a sheet for each participant). Sheet names are the participant’s IDs. However, the IDs are generated in an ID column for all files for more convenience. The following are the excel sheet columns:

  • Column A: ID - Anonymized Id of the user.
  • Column B: Start time - Event start time.
  • Column C: End time - Event start time.
  • Column D: Duration - Duration of the event.
  • Column E: Date - Date of data collection.
  • Column F: Stress level - Reported stress level by the nurse.

Nurses' responses regarding the nature of the stress.

  • Column G: COVID Related
  • Column H: Treating a COVID patient
  • Column I: Patient in Crisis
  • Column J: Patient or patient’s family
  • Column K: Doctors or colleagues
  • Column L: Administration, lab, pharmacy, radiology, or other ancillary services
  • Column M: Increased Workload
  • Column N: Technology related stress
  • Column O: Lack of supplies
  • Column P: Documentation
  • Column Q: Safety (physical or physiological threats)
  • Column R: Lack of supplies
  • Column S: Work Environment - Physical or others: work processes or procedures
  • Column T: Description

Funding

National Science Foundation, Award: 1650551

National Science Foundation, Award: CNS-1429526

Louisiana Board of Regents, Award: LEQSF (2019-20)-ENH-DE-22