A two-stage model using a 345M-parameter pre-trained PPG foundation model plus an attention LSTM predicts in-hospital cardiac arrest from one hour of finger PPG with 0.79 AUROC over the next 24 hours.
Signal quality assessment of ppg signals using stft time-frequency spectra and deep learning approaches
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Continuous Cardiac Arrest Prediction in ICU using PPG Foundation Model
A two-stage model using a 345M-parameter pre-trained PPG foundation model plus an attention LSTM predicts in-hospital cardiac arrest from one hour of finger PPG with 0.79 AUROC over the next 24 hours.