Blood pressure corrected evoked flow responses Blood pressure corrected evoked flow responses Table 3 Percentage of parameters that become significantly different after blood parameters that become significantly different after bloodpressure correction based on 95 % confidence bounds of parameter estimation. N=71 on 95 % confidence bounds of parameter estimation. N=71Percentage parameters changed after blood Percentage parameters changed after blood pressure correction pressure correction mean CBFV systolic CBFV mean CBFV systolic CBFV 28 % 15 % rad/s! 28 % 15 % 33 % 17 % 33 % Tv s17 % 25 % 18 % 25 % K 18 % 41 % 31 % 41 % 31 % Parameter precision was expressed by the relative confidence bound width CP. th s expressed by the relative confidence bound width CP. Pfor each of the 4 percentile) values of Cth - 75Table 4 shows the median (25 an (25th - 75th percentile) values of CPfor each of the 4 values after blood pressure correction are signifi-PNVC model parameters. C . CP values after blood pressure correction are signifi-cantly lower (p<0.005) for all parameters for both mean and systolic data which or all parameters for both mean and systolic data whichmeans that blood pressure correction reduces parameter uncertainty. So, despite re correction reduces parameter uncertainty. So, despitethe fact that the group averaged parameter values hardly differ between uncor- veraged parameter values hardly differ between uncor-rected and corrected estimation, parameters are estimated more precisely when imation, parameters are estimated more precisely whenblood pressure correction is performed. n is performed.In case the confidence bounds of a parameter enclose zero value, the precision of unds of a parameter enclose zero value, the precision ofthe parameter estimate is so poor that basically the parameter cannot be esti- is so poor that basically the parameter cannot be esti-mated and should be considered unreliable. Table 5 shows the percentage of nsidered unreliable. Table 5 shows the percentage ofunreliable parameters without and with blood pressure correction. For the ithout and with blood pressure correction. For the determined on the mean CBFV response a sig-v and T,!parameter values of and Tv determined on the mean CBFV response a sig-nificantly lower fraction (p<0.01) of unreliable parameters occurs with compared (p<0.01) of unreliable parameters occurs with comparedto without blood pressure correction. For K the improvement is less pronounced re correction. For K the improvement is less pronouncedpossibly due to the already narrow confidence bounds listed in table 4. dy narrow confidence bounds listed in table 4. Table 4 Distribution of the%CP values of each NVC model parameter before and f the%CP values of each NVC model parameter before and th and 75th percentile values. N=71, *after correction are displayed as median, 25 played as median, 25th and 75th percentile values. N=71, *p<0.005 p<0.005 Before After Before After After Before After correction correctioncorrectioncorrection correction correction correction (systolic CBFV) (systolic CBFV)(mean CBFV)(mean CBFV) ) (mean CBFV) rad/s 16 % (7-57)(systolic CBFV) 14 % (8-62) 12 % (6-34) *6 % (3-22) *! 6 % (3-22) * 14 % (8-62) 6 % (3-10) * 12 % (7-45) 10 % (6-22) *11 % (7-35) 6 % (3-10) *Tv s 37 % (22-257) 18 % (12-60) * 43 % (23-148) 36 % (19-97) *12 % (7-45) ) 18 % (12-60) * 43 % (23-148) 1 % (1-3) * 2 % (1-6) 1 % (1-4) *2 % (1-11)K 1 % (1-3) * 2 % (1-6) 1 % (1-4) * 91 91
proefschrift gommer inhoud met kaft.indd
To see the actual publication please follow the link above