Eight of $25$ features were high impact variables contributing more than $5\%$ of missing values with the $$red$\_$blood$\_$cells$$ feature being particularly concerning with approximately $15\%$ occurrence of missing values $($Figure $2).$ The random forest and classification tree models exhibited similar accuracy for the imputation techniques tested $($Table $1).$ The mode$-$mean and naive$-$median imputation combinations yielded the highest accuracy for both models, whereas naive$-$mean imputation yielded $93\mathrm{.}8\%$ when tested on the classification tree. There was no difference in model accuracy for the same combination of imputation methods between the random forest and classification tree models.