Table 3.
Studies for ED prediction using the AI framework.
| SN | Citations | IC | DS | GT | Classifier | TOC | ML/DL | ACC % | AUC |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Biswas et al. [182] (2018) | OBBM, LBBM (US) | 407 | Stroke, Diabetes | NR | CNN | DL | 99.61 | 0.99 |
| 2 | Jamthikar et al. [158] (2019) | OBBM, LBBM (US) | 395 | CVD | PCA | RF | ML | 95.00 | 0.80 |
| 3 | Kandha et al. [183] (2020) | OBBM, LBBM | 346 | Death | CNN | NB, SVM, KNN, DT | DL | 83.33 | 0.833 |
| 4 | Jamthikar et al. [160] (2020) | OBBM, LBBM, CUSIP | 202 | CVD | SVM | LR, SVN, ANN |
ML | 92.53 | 0.92 |
| 5 | Saba et al. [184] (2020) | OBBM, LBBM, CUSIP | 246 | Death | 6 Models | SVM | HDL | 89.00 | 0.898 |
SN: serial number, IC: input covariates, DS: data size, GT: ground truth, OBBM: office-based biomarker, LBBM: laboratory-based biomarker, FE: feature extraction, TOC: type of classifier, ACC: percentage accuracy, US: ultrasound, NR: not reported.