. Author manuscript; available in PMC: 2024 Jul 11.

Published in final edited form as: IEEE Access. 2024 Apr 1;12:49122–49133. doi: 10.1109/access.2024.3383789

TABLE 2.

Performance of face detection on USF-MNPAD-I postoperative subjects, subject-wise.

Subject	Images	V5 No Training	V3 Detection	V5 Detection	V6 Detection
A	818	142 (17.3%)	618(75.6%)	652 (79.7%)	818 (100%)
B	1464	0 (0%)	1099 (75.1%)	1427(97.47%)	1464 (100%)
C	294	0 (0%)	294(100%)	294(100%)	294 (100%)
D	3858	3 (~ 0%)	3679(95.4%)	3741(96.9%)	3845 (99.6%)
E	2147	0 (0%)	1994(92.8%)	2147(99.9%)	2147 (100%)
F	505	21 (4%)	504 (99.8%)	476 (94.2%)	505 (100%)
G	3239	136 (4%)	1848 (57%)	3199 (98.7%)	3239 (100%)
H	220	0 (0%)	74 (33.6%)	220 (100%)	220 (100%)

Total	12,545	302 (2.4%)	10,110 (80.5%)	12,156 (96.89%)	12,534 (99.89%)