Fig. 5From: Proposal of a new equation for estimating resting energy expenditure of acute kidney injury patients on dialysis: a machine learning approachA correlation plot between the predicted and observed REE in the test set (20% of unseen data). a Predicted REE by random forest model and b predicted REE by Harris–Benedict equation. We had an r-value of 0.69 with the random forest model (new model) and an r-value of 0.24 with the Harris–BenedictBack to article page