PPG-DaLiA: Heart Rate from Wrist PPG + Accelerometer in Daily Life Activities [15 Subjects, UCI]
Catalog Summary
"Large-scale PPG-based heart rate dataset from 15 subjects wearing wrist (Empatica E4: PPG + 3-axis ACC) and chest (RespiBAN: ECG + respiration + 3-axis ACC) devices during 8 real-life activities. ECG provides HR ground truth. UCI ML Repository. Published in MDPI Sensors 2019 — DOI: 10.3390/s19143079."
Catalog Notes
Overview
PPG-DaLiA (PPG Dataset for motion compensation and Heart Rate Estimation in Daily Life Activities) is a large-scale multimodal dataset created by Attila Reiss, Ina Indlekofer, Philip Schmidt, and Kristof Van Laerhoven at the University of Siegen. It was collected from 15 healthy subjects performing 8 activities representative of daily life under close to real-life conditions: sitting, ascending/descending stairs, table soccer, cycling, driving, lunch break, walking, and working.
Each subject wore two recording devices simultaneously: a wrist-worn Empatica E4 providing PPG (64 Hz) and 3D accelerometer (32 Hz) data, and a chest-worn RespiBAN providing ECG (700 Hz), respiration (700 Hz), and 3D accelerometer (700 Hz). The ECG-derived heart rate from the RespiBAN serves as the ground-truth reference for the wrist PPG-based HR estimation task. The dataset is stored in Python pickle format (one file per subject) and is available from the UCI Machine Learning Repository.
Column Schema
| Signal / Variable | Device | Description | Sampling Rate |
|---|---|---|---|
| PPG | Empatica E4 (wrist) | Photoplethysmography signal (BVP) for HR estimation. | 64 Hz |
| ACC wrist (x/y/z) | Empatica E4 (wrist) | 3-axis accelerometer for motion artifact reference. | 32 Hz |
| ECG | RespiBAN (chest) | Electrocardiogram for ground-truth HR reference. | 700 Hz |
| RESP | RespiBAN (chest) | Respiratory effort signal. | 700 Hz |
| ACC chest (x/y/z) | RespiBAN (chest) | 3-axis accelerometer from chest device. | 700 Hz |
| activity | Label | Activity label (sitting, walking, cycling, driving, etc.). | — |
Key Statistics
- Subjects: 15 healthy subjects
- Activities: 8 daily life activities
- Wrist device: Empatica E4 (PPG at 64 Hz + ACC at 32 Hz)
- Chest device: RespiBAN (ECG + RESP + ACC at 700 Hz)
- File Format: Python pickle (.pkl), one file per subject
- Published: UCI ML Repository, 2019 — DOI: 10.24432/C5SP55
- Associated paper: MDPI Sensors 2019, DOI: 10.3390/s19143079
Use Cases
- Heart rate estimation from wrist PPG with motion artifact cancellation
- Physical effort monitoring using wrist biosensors during real-life activities
- Multi-modal sensor fusion: PPG + accelerometer + ECG + respiration
- Deep learning benchmarking for wearable HR estimation (Deep PPG, CNN, LSTM)
View Data Structure
To explore column names, data types, and sample rows, visit the official dataset page on UCI.
Preview on UCICite This Dataset
The dataset creators ask users of this dataset to cite the accompanying paper. Use one of the verified formats below.
Reiss, A., Indlekofer, I., Schmidt, P., & Van Laerhoven, K. (2019). Deep PPG: Large-Scale Heart Rate Estimation with Convolutional Neural Networks. Sensors, 19(14), 3079. https://doi.org/10.3390/s19143079
Source metadata: MDPI AG (2019) · DOI: 10.3390/s19143079
Indexed by IoTDataset.com on May 10, 2026
Review the Source Record
Confirm the licence, version, access conditions, file format, and provenance at the source before use.