Table 3.
Performance analysis for the proposed uni-step diagnostic system using ConvLSTM DLM on different sub-datasets.
| Category |
Original images |
AWGN corrupted images |
Blurred images |
S&P noisy images |
Denoised images |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| COVID-19 | Normal | Pneumonia | COVID-19 | Normal | Pneumonia | COVID-19 | Normal | Pneumonia | COVID-19 | Normal | Pneumonia | COVID-19 | Normal | Pneumonia | |
| COVID-19 | 0.94 | 0.03 | 0.03 | 0.89 | 0.04 | 0.07 | 0.87 | 0 | 0.13 | 0.89 | 0.02 | 0.09 | 0.92 | 0.01 | 0.07 |
| Normal | 0.01 | 0.99 | 0 | 0 | 0.99 | 0.01 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 |
| Pneumonia | 0.01 | 0 | 0.99 | 0.15 | 0 | 0.85 | 0.03 | 0 | 0.97 | 0.2 | 0.02 | 0.79 | 0.04 | 0 | 0.96 |
| Accuracy | 96.8 % | 90.9 % | 94.6 % | 89.9 % | 96.8 % | ||||||||||
| Testing time | 8.6 Sec. | 7.3 Sec. | 8.9 Sec. | 8.1 Sec. | 8.6 Sec. | ||||||||||