. 2021 Mar 26;11:6950. doi: 10.1038/s41598-021-85699-0

Table 3.

Confusion matrix (actual versus predicted classes), accuracy, sensitivity, specificity, AUROC of the support vector machine classification on each of 11 models.

Model	Variables	Predicted	Actual classes		Accuracy (%)	Sensitivity (%)	Specificity (%)	AUROC (%)
Model	Variables	Predicted	Healthy	PM	Accuracy (%)	Sensitivity (%)	Specificity (%)	AUROC (%)
1	TEPS _fovea→disc and TEPS _fovea→DPE	Healthy	189	9	85.27	77.50	86.70	82.10
1	TEPS _fovea→disc and TEPS _fovea→DPE	PM	29	31	85.27	77.50	86.70	82.10
2	TEPS _distance and TEPS _fovea→DPE	Healthy	177	9	80.62	77.50	81.19	79.3
2	TEPS _distance and TEPS _fovea→DPE	PM	41	31	80.62	77.50	81.19	79.3
3	TEPS _disc→DPE and TEPS _fovea→DPE	Healthy	190	9	85.6	77.50	87.16	82.33
3	TEPS _disc→DPE and TEPS _fovea→DPE	PM	28	31	85.6	77.50	87.16	82.33
4	TEPS _distance and TEPS _fovea→disc	Healthy	190	14	83.72	65.50	87.1	76.08
4	TEPS _distance and TEPS _fovea→disc	PM	28	26	83.72	65.50	87.1	76.08
5	TEPS _disc→DPE and TEPS _fovea→disc	Healthy	189	9	85.27	77.50	86.70	82.10
5	TEPS _disc→DPE and TEPS _fovea→disc	PM	29	31	85.27	77.50	86.70	82.10
6	TEPS _distance and TEPS _disc→DPE	Healthy	201	21	85.27	47.50	92.20	69.85
6	TEPS _distance and TEPS _disc→DPE	PM	17	19	85.27	47.50	92.20	69.85
7	4 TEPS variables	Healthy	192	9	86.43	77.50	88.07	82.79
7	4 TEPS variables	PM	26	31	86.43	77.50	88.07	82.79
8	AL	Healthy	174	14	77.52	65.00	79.82	72.41
8	AL	PM	44	26	77.52	65.00	79.82	72.41
9	CT	Healthy	165	15	73.64	62.50	75.69	69.09
9	CT	PM	53	25	73.64	62.50	75.69	69.09
10	AL & CT	Healthy	167	10	76.36	75.00	76.61	75.80
10	AL & CT	PM	51	30	76.36	75.00	76.61	75.80
11	All variables	Healthy	204	8	91.47	80.00	93.58	86.47
11	All variables	PM	14	32	91.47	80.00	93.58	86.47

AL axial length, AUROC area under receiver operating characteristic curve, CT choroidal thickness, DPE deepest point of the eyeball, PM pathologic myopia, TEPS tomographic elevation of the posterior sclera.