Skip to main content
. 2023 Mar 13;11(6):837. doi: 10.3390/healthcare11060837

Table 5.

Classification accuracy with different train–test splits.

Features Classifier Train–Test Splits Class Precision Recall F1-Score SEN SPE ACC Avg. ACC
Energy + Correlation + Contrast + Entropy + Convolution Block Random Forest 90–10% COVID-19
Normal
100%
100%
100%
100%
100%
100%
100% 100% 100% 97%
80–20% COVID-19
Normal
99%
91%
91%
99%
95%
96%
91% 99% 96%
70–30% COVID-19
Normal
95%
91%
90%
96%
93%
93%
91% 94% 93%