Table 4 Diagnostic efficacy of joint features combined with machine learning models
From: Research on multi-model imaging machine learning for distinguishing early hepatocellular carcinoma
| Â | Â | Accuracy | |
|---|---|---|---|
Classifier | Â | Training group | Test group |
SVM | Â | Â | Â |
| Â | CT | 0.834 | 0.744 |
| Â | MR | 0.887 | 0.789 |
|  | CT + MR | 0.887 | 0.824 |
KNN | Â | Â | Â |
| Â | CT | 0.721 | 0.697 |
| Â | MR | 0.727 | 0.647 |
|  | CT + MR | 0.774 | 0.711 |
RandomForest | Â | Â | Â |
| Â | CT | 0.981 | 0.743 |
| Â | MR | 0.980 | 0.605 |
|  | CT + MR | 0.985 | 0.765 |
XGBoost | Â | Â | Â |
| Â | CT | 1.000 | 0.744 |
| Â | MR | 1.000 | 0.711 |
|  | CT + MR | 1.000 | 0.882 |
LightGBM | Â | Â | Â |
| Â | CT | 0.815 | 0.744 |
| Â | MR | 0.833 | 0.737 |
|  | CT + MR | 0.835 | 0.824 |
MLP | Â | Â | Â |
| Â | CT | 0.771 | 0.718 |
| Â | MR | 0.800 | 0.658 |
|  | CT + MR | 0.782 | 0.824 |