Table 2.

Results of machine learning algorithms for predicting systolic and diastolic blood pressure

Model	Sample	Train set (80%) SBP/DBP [mmHg]	Test set (20%) SBP/DBP [mmHg]
GBM	CC	0.4/0.2 (1.00/1.00)	4.1/2.4 (0.96/0.98)
	IH	0.6/0.4 (1.00/1.00)	5.8/3.7 (0.92/0.95)
LASSO	CC	3.2/1.7 (0.97/0.99)	5.7/2.3 (0.92/0.98)
	IH	4.6/2.9 (0.95/0.97)	6.9/4.0 (0.88/0.95)

Standard deviation of the difference between predicted and reference values of systolic (SBP) and diastolic (DBP) blood pressure for Gradient Boosting model (GBM) and LASSO regression machine-learning algorithms using the CC and IH data. Results are shown separately for the training and test data sets. They were calculated by combining predictions of individual models from 12 study subjects selected as explained in the text. Numbers in brackets correspond to the Pearson correlation coefficients of predicted vs reference SBP/DBP distributions.