Table 4. Systolic peak detection performance on PPG-CBR responders in tropical condition database.

Test	Algorithm	FP	FN	SE (%)	+P (%)
Before Exercise	Method I	26	33	96.55	97.66
	Method II	0	36	96.06	100.00
	Method III	23	42	95.50	97.88
	Method VI	3	0	100.00	99.72
After Exercise 1	Method I	13	27	98.07	99.00
	Method II	1	15	98.67	99.85
	Method III	17	39	96.90	98.67
	Method VI	2	3	99.83	99.79
After Exercise 2	Method I	45	13	98.86	97.29
	Method II	1	26	97.89	99.89
	Method III	7	43	96.71	99.48
	Method VI	1	2	99.86	99.92
After Exercise 3	Method I	36	20	98.19	98.02
	Method II	0	11	99.09	100.00
	Method III	3	37	97.26	99.76
	Method VI	1	2	99.86	99.91

The PPG signals were collected from 40 heat acclimatised emergency responders for 20 seconds during the 10 minutes break of each exercise (cf. Figure 2). To evaluate the performance of each algorithm in each test, two statistical measures are used: and , where TP is the number of true positives (systolic peak detected as systolic peak), FN is the number of false negatives (systolic peak has not been detected), and FP is the number of false positives (non-systolic peak detected as systolic peak).