Table 3. Performance metrics for 4-s segments.
The the p-values are for the pairwise comparison of the accuracy of the proposed model to each of the other classifiers. The proposed network had a highest accuracy and the results were significant at the 95% confidence level (indicated by a†).
| Method | Public datasets | OHCA datasets | ||||||
|---|---|---|---|---|---|---|---|---|
| Se/Sp | BAC/Acc | PPV/NPV | p-val | Se/Sp | BAC/Acc | PPV/NPV | p-val | |
| Proposed network | 99.7/98.9 | 99.3/99.1 | 96.3/99.9 | - | 99.2/96.7 | 98.0/97.5 | 93.6/99.6 | - |
| Deep learners | ||||||||
| Kiranyaz et al [34] | 99.3/98.4 | 98.9/98.6 | 94.6/99.8 | < .05† | 97.7/93.8 | 95.8/95.1 | 88.4/98.8 | < .05† |
| Zubair et al [56] | 95.6/96.2 | 95.9/96.1 | 87.5/98.7 | < .05† | 93.9/89.4 | 91.7/90.9 | 81.0/96.8 | < .05† |
| Acharya et al [39] | 98.4/98.1 | 98.2/98.2 | 93.6/99.5 | < .05† | 94.7/91.6 | 93.2/92.6 | 84.5/97.3 | < .05† |
| Baseline SVM | ||||||||
| SVMall | 98.5/98.3 | 98.4/98.4 | 94.3/99.6 | < .05† | 91.7/96.0 | 93.9/94.6 | 91.7/96.0 | < .05† |
| SVMsel | 99.2/98.7 | 99.0/98.8 | 95.5/99.8 | < .05† | 93.9/97.1 | 95.5/96.1 | 93.9/97.1 | < .05† |