Table 5 This table displays the outcome of a fivefold cross-validation.

From: Medical intelligence using PPG signals and hybrid learning at the edge to detect fatigue in physical activities

Model

Signals

AUC

F1-measure

Precision

Accuracy

48

PPG

0.603

0.601

0.607

0.605

49

PPG

0.672

0.679

0.692

0.678

41

Heartbeats (HR)

0.875

0.872

0.872

0.872

Breathing (BR)

0.635

0.635

0.635

0.636

HR and BR

0.902

0.900

0.904

0.901

Xception with BILSTM

Heartbeats (HR)

0.896

0.887

0.890

0.892

Breathing (BR)

0.756

0.743

0.748

0.746

HR and BR

0.913

0.908

0.909

0.908

ResNetCNN with BILSTM

Heartbeats (HR)

0.909

0.906

0.907

0.905

Breathing (BR)

0.854

0.850

0.853

0.851

HR and BR

0.918

0.912

0.916

0.914

Back to article page