基于特征类内紧凑性的不平衡医学图像分类方法
|
|
孟元, 张轶哲, 张功萱, 宋辉
|
Imbalanced medical image classification based on intra⁃class compactness of features
|
|
Yuan Meng, Yizhe Zhang, Gongxuan Zhang, Hui Song
|
|
表4 各算法在DermaMNIST数据集上的对比实验结果
|
Table 4 Experimental results of different algorithms on the DermaMNIST dataset
|
|
| 模型 | SMOTE | KMSMOTE | UbKNN | ZC3NC |
|---|
| Balanced ACC | Weighted⁃P | Balanced ACC | Weighted⁃P | Balanced ACC | Weighted⁃P | Balanced ACC | Weighted⁃P |
|---|
| Avg | 57.66% | 74.35% | 56.33% | 73.89% | 60.24% | 74.69% | 60.56% | 75.63% | | ResNet18 | 60.65% | 73.25% | 59.00% | 74.92% | 64.07% | 76.12% | 64.46% | 76.46% | | ResNet50 | 49.87% | 71.89% | 47.15% | 70.71% | 52.94% | 72.16% | 53.10% | 72.39% | | ResNeXt50 | 55.34% | 73.35% | 52.60% | 72.35% | 57.05% | 73.84% | 57.30% | 75.77% | | GoogLeNet | 64.75% | 76.92% | 66.57% | 77.59% | 66.91% | 76.62% | 67.36% | 77.90% |
|
|
|