Table 2.
Main findings of the included studies.
| Author | Type of data | AI/ML intervention | Best model | AUC | Sens | Spec | PPV | NPV | Diag. Acc. | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Nazarudin, et al. (27) | Imaging | 2 automated segmentation models: combination of Otsu’s thresholding and the Chan - Vese method, Otsu’s thresholding. | Chan-Vese + Otsu’s segmentation analysis | NR | NR | NR | NR | NR | Remarkable increase in accuracy |
| 2 | Bharati, et al. (28) | Clinical, and imaging data | Gradient boosting, RF, LR, and LR | Hybrid RFLR | 0.93 | NR | NR | NR | NR | 0.91 |
| 3 | Cahyono, et al. (29) | Imaging | Convolutional Neural Network | CNN | NR | NR | NR | NR | NR | |
| 4 | Castro, et al. (30) | Electronic medical records | Algorithm using Natural language processing and codified data | Algorithm using Natural language processing and codified data | NR | NR | NR | 0.68 | NR | NR |
| 5 | RoyChoudhury, et al. (31) | Metabolomics | PLS-DA | Statistical analysis with PLS-DA | 0.8 | NR | NR | NR | NR | NR |
| 6 | Rodriguez, et al. (32) | Virtually generated clinical data | Bayesian network | Bayesian network | NR | NR | NR | NR | NR | NR |
| 7 | Purnama, et al. (33) | Imaging | Neural Network - LVQ method, K-NN and SVM | SVM | NR | NR | NR | NR | NR | 0.83 |
| 8 | Prapty, et al. (34) | Clinical data | KNN, SVM, Naive Classifier, RF | RF | NR | NR | NR | NR | NR | 0.94 |
| 9 | Chauhan, et al. (35) | Clinical data | KNN, Naïve Bayes Classifier, SVM, Decision tree classifier, LR | Decision Tree Classifier | NR | 0.41 | 0.94 | NR | NR | 0.81 |
| 10 | Lawrence, et al. (36) | Imaging | LDC, KNN, SVM | LDC | NR | 0.91 | 0.95 | NR | NR | 0.93 |
| 11 | Mehrotra, et al. (23) | Clinical data | Multivariate logistic regression, Bayesian Classifier | Bayesian classifier | NR | 0.93 | 0.94 | 0.81 | NR | 0.94 |
| 12 | Matharoo-Ball, et al. (37) | Proteomics | Artificial Neural Network | Artificial Neural Network | NR | NR | NR | NR | NR | 1 |
| 13 | Lehtinen, et al. (38) | Clinical data | TPFFN and SOM | TPFFN | NR | NR | NR | NR | NR | efficiency of 97% |
| 14 | Kumar, et al., 2014 REFID 101 (39) | Imaging | PNN, SVM, RBF | PNN | NR | NR | NR | NR | NR | 0.98 |
| 15 | Madhumitha, et al. (40) | Imaging | SVM, K-NN, LR | Proposed Method (SVM + K-NN + LR) | NR | NR | NR | NR | NR | 0.98 |
| 16 | Ho, et al. (41) | Genetics | SVM, RF, GMM | SVM with 5 and 3-fold cross validation | 1 | 1 | 1 | NR | NR | 1 |
| 17 | Gopalakrishnan, et al. (42) | Imaging | SVM. | SVM | NR | NR | NR | NR | NR | 0.94 |
| 18 | Dong, et al. (43) | Clinical data | Orthogonal PLS-DA | Orthogonal PLS-DA | 0.96 | NR | NR | NR | NR | NR |
| 19 | Deshpande, et al. (44) | Clinical and imaging | SVM | SVM | NR | NR | NR | NR | NR | 0.95 |
| 20 | Denny, et al. (45) | Clinical data and imaging | LR, KNN, CART, RFC, NB, SVM | RFC | NR | 0.74 | 0.98 | NR | NR | 0.89 |
| 21 | Deng, et al. (46) | Imaging | Watershed + Object growing algorithm, Level set method, boundary vector field methiod, fuzzy support vector machine classifier | Watershed + Object growing algorithm | NR | NR | NR | NR | NR | NR |
| 22 | Dapas, et al. (47) | Genome wide association | SVM, RF, GMM | NR | NR | NR | NR | NR | NR | NR |
| 23 | Che, et al. (48) | Genetics | Unsupervised hierarchical clustering analysis | Unsupervised hierarchical clustering analysis | NR | NR | NR | NR | NR | NR |
| 24 | Cheng, et al. (49) | Imaging | Gradient boosted trees, Rules based classifier | Rules-based classifier | NA | 0.97 | 0.98 | 0.95 | 0.99 | 0.98 |
| 25 | Zhang, et al. (50) | Clinical data | K-NN, RF, XGB, Stacking classification model | K-NN with follicular fluid | NR | 0.87 | 0.90 | NR | NR | 0.88 |
| 26 | Xie, et al. (51) | Genetics | Random Forest, Artificial Neural Network | Artificial Neural Network | 0.73 | 0.73 | 0.75 | NR | NR | NR |
| 27 | Thakre, et al. (52) | Clinical data | RF, SVM, LR, Gaussian Naïve Bayes, K-NN | RFC | 0.89 | 0.97 | 0.8 | 0.89 | 0.94 | 0.91 |
| 28 | Vikas, et al. (53) | Clinical data | Frequent item set mining, Apriori algorithm | NR | NR | NR | NR | NR | NR | NR |
| 29 | Setiawati, et al. (54) | Imaging | LR, SVM, Backpropagation Neural Network | Backpropagation Neural Network | NR | NR | NR | NR | NR | NR |
| 30 | Rihana, et al. (55) | Imaging | SVM | SVM | NR | 0.88 | 0.95 | NR | NR | 0.9 |
| 31 | Deng, et al. (56) | Imaging | Clustering analysis, Manual image reading | Clustering analysis | 0.84 | NR | NR | NR | NR | 0.84 |
Studies presented by lead author and year of publication with corresponding main findings. Shorthand denoted as: No Response (NR), K-Nearest Neighbor (K-NN), learning vector quantization (LVQ), logistic regression (LR), not reported (NR), support vector machine (SVM), partial least squares discriminant analysis (PLS-DA), topology-preserving feed-forward network (TPFFN), extreme gradient boosting (XGB), self-organizing map (SOM). Classification and Regression Trees (CART), Random Forest (RF), Random Forest Classifier (RFC), Naïve Bayes Classifier (NB), Gaussian mixed model (GMM), Linear Discriminant Classifier (LDC), Convolutional Neural Network (CNN), Random Forest and Logistic Regression (RFLR)