Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

mfARI parameters.

Characterization of a typical model's CBFV response (dotted line) to a BP negative step (dashed line). Δτ is the duration of the transient response. Vbase is the baseline level, which is used to normalize the signal before determining the slope of the straight line that characterize its transient response (thick solid line). The horizontal thick solid straight line is the representation of the constant steady state response (Ks). The angle between the line of transient CBFV response and line of BP recovery is the parameter ϕ. eT and eS are examples of the errors in the transient and steady state phase respectively, their sum is minimized to fit the best set of straight lines.

More »

Fig 1 Expand

Table 1.

Grand mean ± SD of physiological variables, TF coefficients and predicted dynamic CA index values in baseline and hypercapnia.

More »

Table 1 Expand

Fig 2.

Signals in spontaneous fluctuations.

Recordings of spontaneous BP (solid line, mmHg), CBFV (dotted line, cm/s), and EtCO2 (dashed line, mmHg) from a female 29-years-old volunteer, in baseline (A) and hypercapnia (B). An increase in all signals can be observed in the latter after 50 seconds.

More »

Fig 2 Expand

Fig 3.

Mean step responses.

Each curve shows the averaged step responses generated for each subject by NAR models (solid line) and NFIR models (long dashed line) in baseline, and NAR models (dashed line) and NFIR models (dotted line) in hypercapnia. Error bars represent ± SD.

More »

Fig 3 Expand

Fig 4.

Comparison of the actual and predicted CBFV signals.

The CBFV signal predicted by a non-linear NAR model (dotted line) compared with the actual CBFV signal (solid line) of a 26-years-old male volunteer, in baseline (A) and hypercapnia (B).

More »

Fig 4 Expand

Table 2.

ν-SVR hyper-parameters and validation correlations coefficients exhibited by the models selected for both non-linear structures in both conditions.

More »

Table 2 Expand

Fig 5.

ROC curves.

In subplot (A) ROC curves achieved by non-linear models with classic ARI values for NFIR (dotted line) and NAR models (dashed line), and with mfARI values for NFIR (long dashed) and NAR models (solid line). ROC curves for TF are presented in subplot (B) with TF gainVLF (dotted line), TF phaseLF (dashed line), TF gainLF (dot dashed line) and TF phaseVLF (solid line).

More »

Fig 5 Expand

Table 3.

AUC values obtained by each approach when discriminating between baseline and hypercapnia.

More »

Table 3 Expand