Descriptor: Invasive Blood Pressure With Multimodal Physiological Signals During Ergometer Exercise Dataset (ErgoBP)

Continuous physiological time series from wearable devices are widely used in research, aided by the growth of open data sharing. Open datasets rarely pair wearable-derived waveforms with simultaneous ground-truth invasive blood pressure measurements during motion and hemodynamic stress. The ErgoBP dataset comprises recordings from 25 healthy adults who performed moderate- to high-intensity exercise on a cycle ergometer in two body positions. For each participant, approximately one hour of data was collected, spanning rest, graded exercise, and recovery in each posture, encompassing diverse cardiovascular mechanics. Measurements include invasive radial artery blood pressure, photoplethysmography, electrocardiography, wrist accelerometry, and radial tonometry. The dataset includes segmented waveforms, summary measures, quality indices, and auxiliary scripts to aid preprocessing and analysis. The inclusion of invasive blood pressure during exercise in multiple body positions distinguishes ErgoBP from other openly available datasets.