Table 7 Comparative examination of the performance of PulmoNet with other selected models
From: PulmoNet: a novel deep learning based pulmonary diseases detection model
Models | 4 class | 3 class | 2 class (Covid_19 and Healthy) | 2 classes (Pneumonia and Healthy) | No of samples | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Metrics | ACC | PREC | REC | F1 SCORE | ACC | PREC | REC | F1 SCORE | ACC | PREC | REC | F1 SCORE | ACC | PREC | REC | F1 SCORE | BACT_PNEUMONIA | COVID_19 | HEALTHY | VIRAL PNEUMONIA |
(Khan et al. [53] | 89.6 | 90 | 89.92 | 89.8 | 95 | 95 | 96.9 | 95.6 | 99 | 98.3 | 99.3 | 98.5 | NA | NA | NA | NA | 330 | 284 | 310 | 327 |
(Hussain et al. [59] | 91.2 | 91.13 | 90.61 | 92.17 | 94.2 | 93.82 | 93.82 | 94.67 | 99.1 | 98.76 | 97.66 | 97.93 | NA | NA | NA | NA | 400 | 500 | 800 | 400 |
PulmoNet | 94 | 90.52 | 95.01 | 88.65 | 95.4 | 95.15 | 98.54 | 97.34 | 99.4 | 98.86 | 99.41 | 98.46 | 98.3 | 99.12 | 99.01 | 99.24 | 880 | 3,749 | 10,320 | 1,478 |