Estimation of Pulse Transit Time From Radial Pressure Waveform Alone by Artificial Neural Network

Objective: To validate the feasibility of the estimation of pulse transit time (PTT) by artificial neural network (ANN) from radial pressure waveform alone. Methods: A cascade ANN with ten-fold cross validation was applied to invasively and simultaneously recorded aortic and radial pressure waveforms during rest and nitroglycerin infusion (n = 62) for the estimation of mean and beat-to-beat PTT. The results of the ANN models were compared to a multiple linear regression (LR) model when the features of radial arterial pressure waveform in time and frequency domains were used as the predictors of the models. Results: For the estimation of mean PTT and beat-to-beat PTT by ANN (PTT
_ANN
), the correlation coefficient between the PTT
_ANN
and the measured PTT (PTT
_measured
) (mean: r = 0.73; beat-to-beat: r = 0.85) is higher than that between the PTT estimated by LR (PTT
_LR
) and PTT
_measured
(mean: r = 0.46; beat-to-beat: r = 0.79). The standard deviation (SD) of the difference between the PTTANN and PTT
_measured
(mean: SD = 10 ms; beat-tobeat: SD = 8 ms) is significantly less than that between the PTT
_LR
and PTT
_measured
(mean: SD = 17 ms; beat-tobeat: 10 ms), but no significant difference exists between their mean (p = 0.92). The lack of frequency features of radial pressure waveform caused obvious reduction in the correlation coefficient (Δr = 0.16) and SD of the difference (ΔSD = 6 ms) between the PTT
_ANN
and PTT
_measured
. The performance of the ANN was improved by increasing the sample number but not by increasing the neuron number. Conclusion: ANN is a potential method of PTT estimation from a single pressure measurement at radial artery.