BIDMC PPG and Respiration Dataset

@article{pollard_physionet_2026, title = {{PhysioNet} as a global platform for biomedical research}, volume = {1}, issn = {3005-0693}, url = {https://doi.org/10.1038/s44360-026-00096-z}, doi = {10.1038/s44360-026-00096-z}, abstract = {Twenty-five years after its first public repository was established, the PhysioNet resource for biomedical and clinical data has enabled tens of thousands of research papers. Further efforts are ongoing to advance secure data sharing, community engagement and scalable research infrastructure.}, number = {8}, journal = {Nature Health}, author = {Pollard, Tom and Moody, Benjamin E. and Lehman, Li-wei H. and Gow, Brian J. and Fernandes, Chrystinne and Xie, Chen and Johnson, Alistair and Mark, Roger G. and Heldt, Thomas}, month = aug, year = {2026}, pages = {792--795},}

Introduction

This dataset contains signals and numerics extracted from the much larger MIMIC II matched waveform Database, along with manual breath annotations made from two annotators, using the impedance respiratory signal.

Data Collection

The original data was acquired from critically-ill patients during hospital care at the Beth Israel Deaconess Medical Centre (Boston, MA, USA). Two annotators manually annotated individual breaths in each recording using the impedance respiratory signal. The 53 recordings within the dataset, each of 8-minute duration, each contain:

• Physiological signals, such as the PPG, impedance respiratory signal, and electrocardiogram (ECG). These are sampled at 125 Hz.
• Physiological parameters, such as the heart rate (HR), respiratory rate (RR), and blood oxygen saturation level (SpO2). These are sampled at 1 Hz.
• Fixed parameters, such as age and gender
• Manual annotations of breaths.

This dataset was first reported in the referenced publication, in which the data was used to evaluate the performance of different algorithms for estimating respiratory rate from the pulse oximetry, or photoplethysmogram (PPG) signal.

Data Files

The dataset is distributed in three formats:

1. WFDB (WaveForm DataBase) format, which is the standard format used by PhysioNet.
2. CSV (comma-separated-value) format
3. Matlab (r) format, in a manner which is a compatible with the RRest Toolbox of respiratory rate algorithms

WFDB Format

Five files are provided for each recording (where ## is the subject number):

• bidmc##.breath: Manual breath annotations
• bidmc##.dat: Waveform data file
• bidmc##.hea: Waveform header file
• bidmc##n.dat: Numerics data file
• bidmc##n.hea: Numerics header file

Further details on the contents of each file are provided here

CSV Format

Separate CSV files are provided for each recording (where ## is the subject number), containing:

• bidmc_##_Breaths.csv: Manual breath annotations
• bidmc_##_Signals.csv: Physiological signals
• bidmc_##_Numerics.csv: Physiological parameters
• bidmc_##_Fix.txt: Fixed variables

Matlab (r) Format

The *bidmc_data.mat* file contains the following subset of the dataset in a single Matlab (r) variable named *data*. The following are provided for each of the 53 recordings:

• ekg: Lead II ECG signal. Each signal is provided in a structure, where the *v* field denotes the signal values, and *fs* is the sampling frequency.
• ppg: Photoplethysmogram signal
• ref.resp_sig.imp: Impedance respiratory signal
• ref.breaths: Manual annotations of breaths provided by two independent annotators. A vector of sample numbers is provided, which correspond to the signal sample numbers.
• ref.params: Physiological parameters: rr (respiratory rate, derived by the monitor from the impedance signal, breaths per minute), hr (heart rate, derived from the ECG, beats per minute), pr (pulse rate, derived from the PPG, beats per minute), spo2 (blood oxygen saturation level, %).
• fix: A structure of fixed variables, including: id (the MIMIC II matched waveform database subject ID and recording identifier), loc (the ward location), and source (the URLs from which the original data were downloaded).

Cited by

12 publications cite this project. The most recent are listed below.

• IoT-Enabled Healthcare Management: Enhancing Patient Monitoring and Service Delivery Smart Healthcare Management Through Internet of Things Technologies
  Siddhartha Saini, Dr. Ajay Rana, Arun Kumar Rana. Computer and decision making., 2026.
• Context-Augmented, Session-Calibrated Blood Pressure Tracking from Photoplethysmography
  Farouk Ganiyu Adewumi, Timothy Oladunni. arXiv (Cornell University), 2026.
• Detecting Neurovascular Instability from Multimodal Physiological Signals Using Wearable-Compatible Edge AI: A Responsible Computational Framework
  Truong Quynh Hoa, Hoàng Đình Cường, Truong Xuan Khanh. arXiv (Cornell University), 2026.
• High-Altitude Pilot Physiological Monitoring Dataset: Respiratory Performance and SpO₂ Analysis
  Chunyang Jia, Meng Yang, Chuangang Zhao. Scientific Data, 2026.
• An Intelligent IoT and AI Framework for Real-Time Remote Patient Monitoring and Healthcare Management
  Sri Lekha Bandla, Binitkumar M Vaghani. International Journal of Artificial Intelligence and Machine Learning, 2026.

View all 12 publications

Citation data from OpenAlex. Papers that use the project without citing it may be missing.