Ai. Akl et al., Artificial intelligence: A new approach for prescription and monitoring ofhemodialysis therapy, AM J KIDNEY, 38(6), 2001, pp. 1277-1283
The effect of dialysis on patients is conventionally predicted using a form
al mathematical model. This approach requires many assumptions of the proce
sses involved, and validation of these may be difficult. The validity of di
alysis urea modeling using a formal mathematical model has been challenged.
Artificial intelligence using neural networks (NNs) has been used to solve
complex problems without needing a mathematical model or an understanding
of the mechanisms involved. In this study, we applied an NN model to study
and predict concentrations of urea during a hemodialysis session. We measur
ed blood concentrations of urea, patient weight, and total urea removal by
direct dialysate quantification (DDQ) at 30-minute intervals during the ses
sion (in 15 chronic hemodialysis patients). The NN model was trained to rec
ognize the evolution of measured urea concentrations and was subsequently a
ble to predict hemodialysis session time needed to reach a target solute re
moval index (SRI) in patients not previously studied by the NN model (in an
other 15 chronic hemodialysis patients). Comparing results of the NN model
with the DDQ model, the prediction error was 10.9%, with a not significant
difference between predicted total urea nitrogen (UN) removal and measured
UN removal by DDQ. NN model predictions of time showed a not significant di
fference with actual intervals needed to reach the same SRI level at the sa
me patient conditions, except for the prediction of SRI at the first 30-min
ute interval, which showed a significant difference (P = 0.001). This indic
ates the sensitivity of the NN model to what Is called patient clearance ti
me; the prediction error was 8.3%. From our results, we conclude that artif
icial intelligence applications in urea kinetics can give an idea of intrad
ialysis profiling according to individual clinical needs. In theory, this a
pproach can be extended easily to other solutes, making the NN model a step
forward to achieving artificial-intelligent dialysis control. (C) 2001 by
the National Kidney Foundation, Inc.