[Retracted] Atom Search Optimization with the Deep Transfer Learning-Driven Esophageal Cancer Classification Model

Table 2

Result analysis of the ASODTL-ECC technique with various measures.

Label	Accuracy	Sensitivity	Specificity	F-score	MCC	Jaccard index
Entire dataset
Class 0	99.33	99.60	99.20	99.01	98.51	98.03
Class 1	98.27	95.20	99.80	97.34	96.11	94.82
Class 2	98.40	99.20	98.00	97.64	96.46	95.38
Average	98.67	98.00	99.00	98.00	97.02	96.08
Training phase (70%)
Class 0	99.05	99.39	98.89	98.49	97.80	97.02
Class 1	97.71	94.05	99.71	96.67	95.01	93.55
Class 2	97.90	98.86	97.42	96.94	95.39	94.05
Average	98.22	97.44	98.67	97.36	96.07	94.88
Testing phase (30%)
Class 0	100.00	100.00	100.00	100.00	100.00	100.00
Class 1	99.56	98.46	100.00	99.22	98.92	98.46
Class 2	99.56	100.00	99.34	99.33	99.00	98.67
Average	99.70	99.49	99.78	99.52	99.31	99.04