Table 1 Average results of 5-fold cross-validation for SBP and DBP models.

From: Cuff-less blood pressure monitoring via PPG signals using a hybrid CNN-BiLSTM deep learning model with attention mechanism

Architecture

MSE (SBP)

MSE (DBP)

MAE (SBP)

MAE (DBP)

1 CNN + 1 BiLSTM + Attention

13.94

10.66

2.70

2.15

2 CNN + 1 BiLSTM + Attention

12.44

6.76

2.22

1.64

2 CNN + 2 BiLSTM + Attention

9.20

4.50

2

1.40

3 CNN + 2 BiLSTM + Attention

8.62

4.39

1.88

1.34

Back to article page