UCSF ICU bedside-monitor waveforms (private; GPT-PPG / SiamQuality / SQUWA pretraining)
PPGECGResp
Key facts
- Signals
- ppg, ecg, resp
- Reference Labels
- af, alarms, ehr
- Sampling Rates
- native monitor rate not stated in the user papers; GPT-PPG resamples PPG to 40 Hz
- Subjects
- 24100
- License
- Private. UCSF IRB 14-13262 (waiver of consent); used by Emory under a UCSF–Emory DUA. No public release.
Summary
Continuous bedside-monitor waveforms plus arrhythmia alarms and EHR from UCSF Medical Center ICUs, collected March 2013 – December 2018. Used under UCSF IRB 14-13262 (waiver of consent). The Emory authors access it through a UCSF–Emory data-use agreement.
Updated: Oct 11, 2026
Related Research
- GPT-PPG
Foundation model
- SiamQuality
Foundation model
- SiamQuality: A ConvNet-Based Foundation Model for Imperfect Physiological Signals
SiamQuality addresses a critical gap in PPG foundation models: signal quality degradation. Real-worl
- SQUWA: Signal Quality Aware DNN Architecture for Enhanced Accuracy in Atrial Fibrillation Detection from Noisy PPG Signals
Atrial fibrillation (AF), a common cardiac arrhythmia, significantly increases the risk of stroke, h
- A Longitudinal and Multimodal physiological signals dataset of Para handcycling training and race for SCI Rehabilitation
Research dataset
- A Multimodal Dataset for Investigating Working Memory in Presence of Music
Research dataset