Viva Bem Light-to-vigorous physical activities dataset
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Pujolli da Silva, Emely
DOI
Abstract
Description
This dataset consists of synchronized, high-frequency heart rate (HR) time-series data designed to evaluate the accuracy and motion-noise resilience of the Samsung Galaxy Watch 4 (SGW4) smartwatch against the industry-standard Polar H10 ECG chest strap. Collected at a sampling rate of 1 Hz via the Viva Sensing application, the dataset captures broad heart rate dynamics across light, moderate, and vigorous intensities during supervised 10-minute sessions of stationary cycling and treadmill workouts. The target population comprises healthy adult volunteers recruited under approved human research ethics guidelines (CAAE: 55532622.0.0000.5404), ensuring a cohort free of cardiorespiratory conditions or physical limitations that could affect the protocol's execution.To facilitate advanced validation and algorithm benchmarking, the dataset features precise contextual annotations regarding exercise type, target intensity zones calculated using the standard formula, and specific physical constraints. A key characteristic of this database is the inclusion of two distinct experimental scenarios for each activity: one where subjects restricted their arm movements to minimize external noise, and another where they moved naturally. This deliberate variation in motion artifacts makes the "Viva bem database" a valuable asset for researchers in digital health, biomedical engineering, and wearable technology seeking to develop and test filtering algorithms or machine learning models for photoplethysmography (PPG) sensors. Due to privacy and ethical considerations, only a portion of the dataset may be publicly available. Researchers interested in accessing the complete dataset should contact the authors and submit a formal request, which will be verified on a case-by-case basis to ensure compliance with privacy and ethical standards. The PGD is made available here