EnviroWellBeing Dataset

The EnviroWellBeing Dataset was created to fill a critical gap in existing datasets, which often lack detailed, synchronized, real-time measurements of both physiological and environmental factors. This dataset is designed to address complex research questions related to the immediate effects of environmental factors on human wellbeing. By integrating a variety of sensors, this dataset enables the exploration of how different environmental conditions—such as temperature, noise, and air quality—affect human health responses. Dataset Composition: The dataset includes the following components: Physiological Data: Heart rate (HR) Breathing rate (BR) Skin temperature Electrodermal activity (EDA) Posture and activity level High-fidelity electrocardiogram (ECG) for heart rate variability analysis Environmental Data: Ambient temperature Light intensity Noise level Air quality measurements (concentrations of CO, CO2, NO, NO2, SO2, VOCs) Ultraviolet (UV) exposure Air pressure Wristband Sensor Data: Skin body temperature Self-reported stress levels The dataset consists of synchronized time-series data from wearable sensors (BioHarness chest sensor, Microsoft Band 2 wristband) and environmental data loggers, allowing for the study of complex interactions between human health and environmental factors. Data Collection Method: Data was collected through a series of controlled experimental conditions where participants were exposed to varied environmental scenarios while wearing multiple sensors: BioHarness Chest Belt: Measures physiological parameters, including heart rate, skin temperature, and movement. Microsoft Band 2 Wristband: Measures electrodermal activity, heart rate, UV exposure, noise levels, and body temperature. Environmental Data Logger: Records ambient environmental parameters such as temperature, light levels, air pressure, and gas concentrations. Preprocessing: The raw data from the sensors was preprocessed to ensure consistent time-series analysis: Data from different sensors, sampled at varying frequencies (1Hz for the chest sensor, 8Hz for wristband data, and 0.5Hz for the environmental logger), was harmonized to a uniform 1Hz frequency. Missing data was handled through backward filling, and outliers were addressed through histogram analysis. Feature extraction techniques were applied to the ECG data to derive heart rate variability metrics. Potential Research Applications: Human-Environment Interaction: Analyzing how environmental factors like temperature, noise, and pollutants impact physiological responses and overall health. Health Monitoring Systems: Using real-time data to develop systems that alert individuals to environmental factors that may negatively impact their health. Personalized Wellbeing Interventions: Exploring how individualized health interventions can be designed based on real-time environmental data. Environmental Psychology: Studying the psychological effects of environmental factors such as noise pollution and temperature fluctuations.