OBJECTIVE: The aim of this proof-of-concept study was to evaluate whether combining population pharmacokinetic and machine learning approaches could provide a more accurate prediction of the clearance of renally eliminated drugs in individual neonates.
METHODS: Six drugs that are primarily eliminated by the kidneys were selected (vancomycin, latamoxef, cefepime, azlocillin, ceftazidime, and amoxicillin) as 'proof of concept' compounds. Individual estimates of clearance obtained from population pharmacokinetic models were used as reference clearances, and diverse machine learning methods and nested cross-validation were adopted and evaluated against these reference clearances. The predictive performance of these combined methods was compared with the performance of two other predictive methods: a covariate-based maturation model and a postmenstrual age and body weight scaling model. Relative error was used to evaluate the different methods.
RESULTS: The extra tree regressor was selected as the best-fit machine learning method. Using the combined method, more than 95% of predictions for all six drugs had a relative error of < 50% and the mean relative error was reduced by an average of 44.3% and 71.3% compared with the other two predictive methods.
CONCLUSION: A combined population pharmacokinetic and machine learning approach provided improved predictions of individual clearances of renally cleared drugs in neonates. For a new patient treated in clinical practice, individual clearance can be predicted a priori using our model code combined with demographic data.
RESULTS: We found that none of the monosaccharides that make up the plant cell wall polysaccharides specifically inhibit Salmonella attachment to the bacterial cellulose-based plant cell wall models. Confocal laser scanning microscopy showed that Salmonella cells can penetrate and attach within the tightly arranged bacterial cellulose network. Analysis of images obtained from atomic force microscopy revealed that the bacterial cellulose-pectin-xyloglucan composite with 0.3 % (w/v) xyloglucan, previously shown to have the highest number of Salmonella cells attached to it, had significantly thicker cellulose fibrils compared to other composites. Scanning electron microscopy images also showed that the bacterial cellulose and bacterial cellulose-xyloglucan composites were more porous when compared to the other composites containing pectin.
CONCLUSIONS: Our study found that the attachment of Salmonella cells to cut plant cell walls was not mediated by specific carbohydrate interactions. This suggests that the attachment of Salmonella strains to the plant cell wall models were more dependent on the structural characteristics of the attachment surface. Pectin reduces the porosity and space between cellulose fibrils, which then forms a matrix that is able to retain Salmonella cells within the bacterial cellulose network. When present with pectin, xyloglucan provides a greater surface for Salmonella cells to attach through the thickening of cellulose fibrils.