Table 5.
Summary of ML models, data modalities, and metrics of the papers reviewed in this survey.
| References | Journal/conference | Data | Data modality | ML | Metrics |
|---|---|---|---|---|---|
| Magesh et al. (2020) | Iternation Journal of Pervasive Computing and Communications | Thermal, Acoustic | 1-2 | RNN, LSTM | N/A |
| Vedaei et al. (2020) | IEEE access | Health parameters | 4 | SVM, Decistion tree | Accuracy: 68.9–76.9%, F1-score: 69.7–77.3% |
| Purnomo et al. (2021) | Sensors | Breathing Movement | 1 | XGBoost, MFCC | Accuracy: 87.38% |
| Almalki et al. (2022) | Computing | UAV Thermal image | 1 | CNN, MANN | Accuracy: 82.63%, F-1 score: 0.98 |
| Alsarhan et al. (2021) | International Journal of Interactive Mobile Technologies | Contact tracing data | 1 | RL | Packet loss probability: 0.1–0.4, Arrival rates: 80–120 |
| Fahad et al. (2022) | Biomedical Engineering: Applications, Basis and Communications | CT images | 2 | AI-PSR model | N/A |
| Barnawi et al. (2021) | Future Generation Computer Systems | UAV Termal image | 2 | CNN, DCNN | Accuracy: 98–99.4%, Precision: 100%, 96–99% |
| Karmore et al. (2022) | IEEE Sensors Journal | Humanoid modules | 6 | Decision tree, TCN | Sensitivity: 95.39%, Specificity: 97.60%, Precision: 95.47%, Accuracy: 97.95% |
| Mir et al. (2022) | Journal of Healthcare Engineering | IoT Sensors | 7 | SVM, decision tree, NB, LR, NN | SVM Accuracy: 93.0% |
| Muhammad et al. (2021) | IEEE Network | Cough sound, Chest X-ray | 2 | FL | Accuracy: 95%, Precision: 97%-99% |
| Khelili et al. (2022) | Biomedical Signal Processing and Control | X-ray images | 3 | CNN | Classification: 97%, Precision: 100% |
| Singh and Kaur (2021) | World Journal of Engineering | Framework measurement | 4 | Fog computing | Classification: 81.2%, Kappa: 0.732, RMSE: 0.241 |
| Alanazi et al. (2020) | Journal of healthcare engineering | COVID-19 data | 3 | Statistic analysis | N/A |
| Zhou et al. (2021) | Applied soft computing | CT images | 2 | CNN, Transfer learning, Ensemble learning | Accuracy: 97–99.05% |
| Shorfuzzaman (2021) | Computing | CT images | 2 | CNN | Accuracy: 96.58%, Precision & Specificity: 99.16%, AUC score: 96.6% |
| Orlandic et al. (2021) | Scientific Data | Audio dataset | 7 | XGB, CV | Precision: 95.4%, Sensitivity: 78.2%, Specificity: 95.3%, Balanced Accuracy: 86.7,% AUC: 96.4% |
| Xia et al. (2021) | NeurIPS | Audio dataset | 6 | SVM, CNN | AUC: 75% Sensitivity: 70%, Specificity: 70% |
| Dang et al. (2019) | Journal of Medical Internet Research | Audio dataset | 3 | GRU | AUC: 79%, Sensitivity: 75%, Specificity 71% |
| Ardabili et al. (2020) | Algorithms | COVID dataset | 2 | MLP, ANFIS | N/A |
| Vekaria et al. (2020) | IEEE Internet of Things Journal | IoT and economy data | 5 | LSTM | MAPE: 1.27%, RMSE: 6308 |
| Wang et al. (2020) | IEEE Access | Social Internet of Things (SIoT) data | 2 | FL, GNN | N/A |