Table 4. Characteristics of deep learning systems for CVD risk predictions (II).
SN | Studies | # GT | GT N | # AU | DL | # ML C | CT | FE | HID | HU | LU | Protocol |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Azzorpardi et al. (78) | 2 | CVD, non-CVD | 1 | UNet | × | × | √ | √ | × | 4 | 15 |
2 | Biswas et al. (79) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | √ | × | 4 | 10 |
3 | Biswas et al. (80) | 2 | CVD, non-CVD | 1 | ANN | × | × | √ | √ | × | 3 | 1 |
4 | Biswas et al. (81) | 2 | CVD, non-CVD | 1 | CNN, FNN | × | × | √ | √ | √ | 10 | 10 |
5 | Jain et al. (98) | 2 | CVD, non-CVD | 1 | UNet | × | × | √ | √ | × | 4 | 10 |
6 | Jain et al. (99) | 2 | CVD, non-CVD | 1 | HDL, SDL | × | × | √ | √ | × | 3 | 10 |
7 | Jamthikar et al. (13) | 3 | CVD, non-CVD, RA | 1 | DL | √ | SVM | √ | √ | × | 1 | × |
8 | Lakadir et al. (82) | 2 | CVD, non-CVD | 1 | CNN | √ | SVM | √ | √ | × | 9 | 5 |
9 | Meshram et al. (83) | 2 | CVD, non-CVD | 1 | UNet | × | × | √ | √ | × | 4 | × |
10 | Wu et al. (84) | 2 | CVD, non-CVD | 1 | DeepMAD | × | × | √ | × | × | 3 | × |
11 | Zhou et al. (85) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | MCar* |
12 | Zhou et al. (86) | 2 | CVD, non-CVD | 1 | Unet++, CNN | × | × | √ | √ | × | 4 | × |
13 | Ganitidis et al. (87) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | √ | × | 4 | 4 |
14 | Mohannadi et al. (100) | 2 | CVD, non-CVD | 1 | CNN, Unet | × | × | √ | × | √ | 6 | × |
15 | Latha et al. (101) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | √ | 3 | × |
16 | Otgonbaatar et al. (102) | 2 | Cerebrovascular Disease | 3 | FBP, HIR, DLR | × | × | √ | × | × | 3 | × |
17 | Jain et al. (103) | 2 | CVD, non-CVD | 2 | SegNet-Unet | × | × | √ | × | √ | 224 | 10 |
18 | Ziegler et al. (88) | 2 | CVD, non-CVD | 1 | CNN, DM | × | × | √ | × | × | 11 | 10 |
19 | Bortsova et al. (89) | 2 | CVD, non-CVD | 1 | DeepEnsemble | × | × | √ | × | √ | 4 | 10 |
20 | Zhu et al. (104) | 2 | CVD, non-CVD | 1 | Unet++ | × | × | √ | × | √ | 2 | 5 |
21 | Park et al. (105) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | √ | × | 10 |
22 | Jain et al. (90) | 2 | CVD, non-CVD | 1 | FNCNN | × | × | √ | × | √ | 3 | 2, 5, 10 |
23 | Savaş et al. (91) | 2 | CVD, non-CVD | 2 | DNN, ANN | × | × | √ | × | 1 | 3 | × |
24 | Washim et al. (92) | 2 | CVD, non-CVD | 1 | CFNet | × | × | √ | × | × | 5 | × |
25 | Sudha et al. (93) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | × |
26 | Groves et al. (106) | 2 | CVD, non-CVD | 1 | R-CNN | × | × | √ | × | × | 2 | 4 |
27 | Saba et al. (94) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | × |
28 | Tsakanikas et al. (107) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | × |
29 | Koktzoglou et al. (95) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | × |
30 | Flores et al. (108) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | 5,10 |
31 | Luo et al. (96) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | × |
32 | Xiao et al. (97) | 2 | CVD, non-CVD | 1 | CNN | × | × | √ | × | × | 4 | 4 |
CVD, cardiovascular disease; SN, serial number; # GT, ground truth; GT N, GT name; # AU, number of algorithm used; DL, deep learning; # ML C, number of ML classifier; CT, classifier type; FE, feature extraction; HID, handling imbalanced data; HU, hyperparameters used; LU, layers used.