DocumentCode :
3342193
Title :
Dose of dialysis predicted by continuous measurement of dialysate urea concentration
Author :
Cappello, A. ; Tartarini, R. ; Paolini, F. ; Calzavara, P.
Author_Institution :
Dipt. di Elettronica, Inf. e Sistemistica, Bologna Univ., Italy
Volume :
5
fYear :
1996
fDate :
31 Oct-3 Nov 1996
Firstpage :
1813
Abstract :
On-line estimation of urea kinetic parameters from dialysate urea concentration (DUN) is proposed and applied to dialysis dose prediction during therapy. Continuous measurement of DUN was performed by a post-filter enzymatic sensor (Hospal Dasco Spa) on six patients undergoing standard haemodialysis. In order to allow for early and reliable prediction of the end of dialysis values, a variable-volume double-pool model is used since the classical first order model cannot account for the fast-to-slow dynamics occurring during the first hour. Three parameter combinations are estimated at each time on the basis of the past DUN history: Kd/V0, KdC0 and VR, where Kd, Vo, Co, and VR are the dialyzer clearance, the total urea distribution volume, the initial plasma concentration, and the intra-extracellular volume ratio, respectively. Residuals between model-predicted and measured data have an average root mean square error of about 3% of the DUN excursion and model parameters show realistic values. The final dose of dialysis is predicted by multiplying Kd/Vo for the scheduled duration of the dialysis session. Prediction after 70 min from the beginning of the session differs on the average by less than 5% from the final value thus demonstrating the applicability of the proposed method. Furthermore, knowledge of model parameters allows for early and accurate prediction of the total urea removed
Keywords :
biomedical measurement; blood; chemical variables measurement; organic compounds; patient treatment; 1 hr; dialysate urea concentration continuous measurement; dialysis dose prediction; initial plasma concentration; intra-extracellular volume ratio; measured data; model parameters; model-predicted data; post-filter enzymatic sensor; residuals; root mean square error; standard haemodialysis; therapy; total urea distribution volume; variable-volume double-pool model; History; Kinetic theory; Measurement standards; Medical treatment; Parameter estimation; Performance evaluation; Plasma measurements; Predictive models; Root mean square; Virtual reality;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE
Conference_Location :
Amsterdam
Print_ISBN :
0-7803-3811-1
Type :
conf
DOI :
10.1109/IEMBS.1996.646268
Filename :
646268
Link To Document :
بازگشت