Description
We propose a multimodal physiological signal dataset called LTHRV, which contains high-quality RGB video, PPG signals, and breathing signals. Data were collected in a controlled experimental setting, with subjects in a supine position performing two different experimental conditions: natural breathing and apnea, each with a recording duration of 15 minutes. Among them, under natural breathing conditions, subjects maintained independent and even breathing; In apnea conditions, subjects hold their breath briefly at a set time point to enhance heart rate fluctuations and increase the range of heart rate variability (HRV).
Dataset Details
Subjects: 18 subjects (12 male, 6 female) with age range 21 to 25 years
Camera: RGB Logitech C920 camera - 30fps, 640x480 pixels stored in an uncompressed YUV planar pixel format (approx. 15GB per video)
Reference signal: Contec CMS50EA pulse oxygen meter + HKH-11C breathing sensor (synchronous collection of cardiopulmonary signals)
Activities:
- Natural breathing: Participants lie flat and keep breathing normally, and the recording duration is 15 minutes.
- Apnea: At the 5th and 10th minutes, participants held their breath (BH) for 20 seconds each and kept breathing normally for the rest of the time.
Download
Download the request form, print it, sign it, and send it to rcsong@hfut.edu.cn. Once received, we will share a download link. The dataset size is approximately 540 GB. It is distributed in a single file compressed with a 7zip program.
Manual
RawData/
||---apnea/
||------s*/
||----------spo2/
||--------------wave.csv
||----------resp/
||--------------resp.csv
||----------video.avi
||---normal/
||------s*/
||----------spo2/
||--------------wave.csv
||----------resp/
||--------------resp.csv
||----------video.avi
where * is the subject ID (* ∈ {1,2,...,15} ), and the data is categorized into Apnea and Normal experimental conditions. Each subject's data includes video.avi (RGB video), wave.csv (blood volume pulse signal), and resp.csv (respiratory signal).
Citation
If you use this dataset in your research, please cite the following publication: