Table 1.
Performance of machine learning models for disease detection using biomarker and clinical outcomes (2018–2023).
| Machine Learning | Studies | Outcome | Ref. |
|---|---|---|---|
| Deep learning system | Diabetic & cancer | Specificity: <96 %, sensitivity: <87 % | [82] |
| ANN | Lung cancer | Specificity: 96 %, sensitivity: 95.8 % | [83] |
| Naïve bayes and RF classifier | Blood cancer | Accuracy: 96.6 % | [84] |
| ANN | Breast cancer | Accuracy: 90 % | [85] |
| Gradient boosting | COPD | Accuracy: 91.3 %, sensitivity: 100 % | [86] |
| Gaussian Naïve Bayes classifier and SVM | Stroke | R2: 0.97 | [87] |
| Decision support system | Dementia | AUC for DSI: 0.79; 0.75 | [37] |
| -Designed unsupervised learning | Dementia | P-value: 0.024; 0.018 | [13] |
| SVM classifier | CVD | Accuracy: 90.2 % | [75] |
| Feature extraction | oHCM | Accuracy: 99 % | [39] |
| ANN | Cardio. | Accuracy:>98 % | [88] |
| SVM and CNNs | Liver | – | [89] |
| Mann-Whitney-RF technique | Liver toxicity | FI: 0.91 | [90] |
| Network-based prediction | COVID-19 | – | [81] |