CapnoBase IEEE TBME RR benchmark (Karlen 2013)
The CapnoBase TBME RR benchmark contains 42 eight-minute recordings with PPG and respiration, used to compare respiratory rate algorithms on the same data. The authors ask that it not be used to train or tune algorithms, because that would bias the comparison.
- Published: Borealis, 2021
- Size: 42 cases, 8 minutes each
- Use: test-only benchmark for respiratory rate algorithms
- License: custom terms specific to this dataset
- Type
- Dataset
- Host
- Not reported
- Licence
- Custom terms specific to this dataset
- Access
- Not reported
- Subjects
- Not reported
- PPG sites
- Not reported
- Setting
- Clinical
- Wrist
- Not reported
- Raw waveforms
- Not reported
- Tasks
- HR estimation
- Updated
- Oct 16, 2026
Original description
The CapnoBase TBME RR benchmark dataset contains 42 cases of 8-min recordings. In addition to the CO2 waveforms (capnograms), these cases have also the Photoplethysmogram (PPG) from pulse oximetry available. Labels from an expert are available for pulse peaks from PPG and breaths from CO2. Also, the benchmark contains the results from the Karlen et al. IEEE TBME paper on multi-parameter RR calculation. This can be used to directly compare algorithm performances. The benchmark dataset is used by researchers to test and compare algorithms. This dataset should not be used to train or tune an algorithm as it may bias the performance results.
Verbatim fromDataCite· Licence: info:eu-repo/semantics/openAccess
How it works
No model in the hub reports results on this dataset yet.
Models trained or evaluated on it
Cite
@misc{capnobase2026,
title = {{CapnoBase IEEE TBME RR benchmark (Karlen 2013)}},
year = {2026},
doi = {10.5683/SP2/NLB8IT},
url = {https://doi.org/10.5683/SP2/NLB8IT}
}History
Record updated