The synthetic dataset is a collection of simulated electrocardiogram (ECG) and pulse oximetry (photoplethysmogram, PPG) signals. The signals were generated to allow researchers to check that their algorithms are able to estimate respiratory rate (RR) accurately on idealised data. It is particularly helpful for ensuring that algorithms are reasonably implemented.
The dataset contains ECG and PPG signals across covering a range of heart rates (HRs) and RRs. Three signals are provided for each combination of HR and RR, each exhibiting either baseline wander (BW), amplitude modulation (AM), or frequency modulation (FM).
The dataset is provided in Matlab format. Firstly, the source code used to generate the dataset is provided (here). This allows users to modify the dataset to their own liking. Secondly, the dataset is provided having used a particular configuration of the source code to produce it (below).
This dataset was first reported in this publication, which contains additional information on the dataset.
When using this dataset please cite this publication:
Charlton P.H. and Bonnici T. et al. An assessment of algorithms to estimate respiratory rate from the electrocardiogram and photoplethysmogram, Physiological Measurement, 37(4), pp. 610-26, 2016
Link to downloadThe dataset is provided in three formats. Use the links below to download the dataset in your preferred format:
- Matlab ® Format
- CSV Format: Comma-Separated Value format.
- WFDB Format: WaveForm DataBase format, as used in .
- License File
- ReadMe File: Viewable in any text reader, although for easier reading see the formatted version here.
- Script for generating dataset: The source code used to generate the dataset is provided, allowing users to modify the dataset to their own liking (Matlab ® file).
- Script for converting dataset to other formats: The source code used to convert the dataset into CSV and WFDB formats is provided.