Research Article
Alzheimer’s Disease Classification Based on Image Transformation and Features Fusion
Table 3
Experimental results of multimodal data fusion.
| Methods | Criteria | NC/SMC | NC/MCI | SMC/MCI | SMC/AD | MCI/AD | NC/AD | EMCI/LMCI |
| mALFF+mReHo+improved 3DPCANet+tandem +SVM | Accuracy | 73.91% | 81.48% | 85.00% | 94.44% | 81.82% | 80.00% | 70.37% | Sensitivity | 80.00% | 83.33% | 91.67% | 100.00% | 100.00% | 73.33% | 64.71% | Specificity | 62.50% | 80.00% | 75.00% | 90.00% | 60.00% | 90.00% | 80.00% | F1 | 80.00% | 80.00% | 88.00% | 94.12% | 85.71% | 81.48% | 73.33% | AUC | 61.70% | 62.22% | 89.58% | 95.00% | 75.00% | 76.70% | 84.62% | mALFF+mReHo+improved 3DPCANet+CCA+softmax | Accuracy | 73.91% | 74.07% | 85.00% | 77.78% | 77.27% | 76.00% | 77.78% | Sensitivity | 86.67% | 83.33% | 91.67% | 75.00% | 91.67% | 80.00% | 100.00% | Specificity | 50.00% | 66.67% | 75.00% | 80.00% | 60.00% | 70.00% | 40.00% | F1 | 81.25% | 74.07% | 88.00% | 75.00% | 81.48% | 80.00% | 85.00% | AUC | 68.33% | 75.00% | 83.33% | 77.50% | 75.83% | 75.00% | 70.00% | mALFF+mReHo+improved 3DPCANet+CCA+SVM | Accuracy | 91.30% | 88.89% | 95.00% | 83.33% | 86.36% | 92.00% | 85.19% | Sensitivity | 93.33% | 83.33% | 91.67% | 75.00% | 91.67% | 93.33% | 94.12% | Specificity | 87.50% | 93.33% | 100.00% | 90.00% | 80.00% | 90.00% | 70.00% | F1 | 93.33% | 86.96% | 95.65% | 80.00% | 88.00% | 83.33% | 88.89% | AUC | 81.70% | 82.22% | 92.71% | 82.50% | 85.83% | 91.30% | 82.06% |
|
|