Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence

Table 6 AI model prediction performance using training data without Braden metrics

AI technique	Original (row, imbalanced) data			Post-processed (rebalanced) cohorts
LASSO linear regularized model	Model			Model

	Performance of testing data			Performance of testing data
	Confusion matrix	Result false	Result true	Confusion matrix	Result false	Result true
	Prediction False	4775	150	Prediction False	4111	48
	Prediction True	21	67	Prediction True	685	169
	Accuracy	96.59%		Accuracy	85.38%
	95% CI	96.05%	97.07%	95% CI	84.37%	86.35%
	Kappa	0.425		Kappa	0.265
	Sensitivity	0.30876		Sensitivity	0.7788
	Specificity	0.99562		Specificity	0.8571
	Prevalence	0.04329		Prevalence	0.04329
Random forest

ISSN: 1472-6947