A Computer-Aided Heart Valve Disease Diagnosis System Based on Machine Learning

Table 3

(a). Model assessment indicators of the trained GoogLeNet model. (b) Model assessment indicators of the trained WKNN model.

Indicators	AS (%)	MR (%)	MS (%)	MVP (%)	N (%)
(a)
Accuracy	99.00	99.00	99.50	97.50	100.00
Sensitivity	100.00	97.50	100.00	90.00	100.00
Specificity	98.75	99.38	99.38	99.38	100.00
F1 score	99.37	98.43	99.69	94.46	100.00
(b)
Accuracy	98.50	91.50	95.00	93.50	99.50
Sensitivity	95.00	77.50	85.00	87.50	100.00
Specificity	99.38	95.00	97.50	95.00	99.38
F1 score	97.14	85.36	90.82	91.10	99.69