Table 6.

The 11 nonlinear trimodal regression analysis parameters were used to assess cardiovascular risks on subjects grouped by age through machine learning algorithms and evaluation metrics were computed.

	Age [years]	Algor.	Acc. Mean [%]	Acc. Max [%]	Sens. [%]	Spec. [%]	Recall [%]	Precision [%]	AUCROC
CHD	66–75	GB	79.0	81.7	73.6	84.5	73.6	82.6	0.883
		RF	87.5	90.2	83.7	91.3	83.7	90.5	0.953
		ADA-B	84.3	89.3	79.5	89.2	79.5	88.0	0.920
	76–83	GB	75.9	79.7	69.5	82.3	69.5	79.7	0.862
		RF	85.3	87.4	82.9	87.7	82.9	87.1	0.930
		ADA-B	77.0	82.3	71.0	83.0	71.0	80.7	0.836
	84–98	GB	72.8	81.1	79.4	66.2	79.4	70.2	0.825
		RF	82.0	86.2	86.3	77.7	86.3	79.4	0.908
		ADA-B	71.3	81.0	81.0	61.7	81.0	61.7	0.769
CVD	66–75	GB	76.4	79.6	70.9	81.9	70.9	79.6	0.868
		RF	85.4	88.0	81.7	89.1	81.7	88.2	0.937
		ADA-B	78.3	82.1	73.6	83.1	73.6	81.3	0.858
	76–83	GB	72.0	76.2	66.4	77.6	66.4	74.8	0.817
		RF	80.6	82.4	78.5	81.9	78.5	82.7	0.890
		ADA-B	66.1	73.2	61.1	71.2	61.1	68.0	0.725
	84–98	GB	68.9	76.2	77.2	60.6	77.2	66.2	0.786
		RF	78.0	88.9	83.7	72.4	83.7	75.2	0.875
		ADA-B	62.9	67.6	64.3	61.5	64.3	62.5	0.659
CHF	66–75	GB	93.8	95.0	91.5	96.0	91.5	95.8	0.981
		RF	97.9	99.2	97.5	98.4	97.5	98.4	0.998
		ADA-B	97.5	98.7	96.2	98.8	96.2	98.8	0.997
	76–83	GB	88.1	90.5	84.6	91.7	84.6	91.0	0.963
		RF	96.0	97.0	94.7	97.4	94.7	97.3	0.995
		ADA-B	94.2	96.4	92.2	96.1	92.2	95.9	0.986
	84–98	GB	82.6	87.8	88.2	76.9	88.2	79.2	0.921
		RF	92.6	96.4	92.8	92.4	92.8	92.4	0.981
		ADA-B	89.9	93.5	92.5	87.2	92.5	87.9	0.964