Dataset Hub

The CIME PPG dataset is described in two papers by Ke Xu et al., serving as training and testing sets respectively. It is primarily used for research on motion‑artifact removal and pulse‑rate variability extraction from photoplethysmography (PPG) signals.

The TROIKA dataset was introduced by Zhang and collaborators in a 2015 IEEE Transactions on Biomedical Engineering paper for heart rate monitoring. It contains photoplethysmography (PPG) and tri‑axial accelerometer signals from 12 male participants aged 18–35. Data were collected during rest, walking, running, and cooling phases, sampled at 125 Hz, and include reference heart rate values derived from ECG signals as ground truth.