Optimal Deep Learning-Based Vocal Fold Disorder Detection and Classification Model on High-Speed Video Endoscopy

Table 1

Result analysis of ODL-VFDDC technique under distinct parameters and runs.

Parameters	Accuracy	Sensitivity	Specificity
Run-1
Initial frequency (Fo)	58.49	57.30	61.86
Jitter (%)	77.23	97.20	16.71
Shimmer (%)	79.00	93.61	34.15
HNR (dB)	64.66	71.46	55.02
Average	69.85	79.89	41.94
Run-2
Initial frequency (Fo)	60.21	59.49	62.01
Jitter (%)	77.97	97.62	15.79
Shimmer (%)	76.48	95.67	34.67
HNR (dB)	68.03	71.30	57.18
Average	70.67	81.02	42.41
Run-3
Initial frequency (Fo)	58.16	58.11	60.68
Jitter (%)	77.69	96.82	18.37
Shimmer (%)	78.56	95.45	34.08
HNR (dB)	65.95	70.20	55.40
Average	70.09	80.15	42.13
Run-4
Initial frequency (Fo)	58.42	57.92	61.09
Jitter (%)	78.47	99.09	17.18
Shimmer (%)	76.89	93.52	35.10
HNR (dB)	64.65	68.84	56.84
Average	69.61	79.84	42.55
Run-5
Initial frequency (Fo)	59.21	57.71	62.71
Jitter (%)	77.79	98.05	17.87
Shimmer (%)	77.75	94.03	34.37
HNR (dB)	66.10	70.35	56.56
Average	70.21	80.04	42.88