Table 5 Results of our CNN models against other methods
From: Intelligent diagnosis with Chinese electronic medical records based on convolutional neural networks
Model | Precision(%) | Accuracy(%) | F1-score(%) |
|---|---|---|---|
1-layer CNN | 83.94 | 83.72 | 83.78 |
1-layer LSTM | 43.97 | 46.33 | 38.18 |
1-layer GRU | 82.95 | 82.2 | 82.37 |
2-layers CNN | 82.81 | 82.58 | 82.65 |
2-layers LSTM | 23.01 | 34.12 | 19.57 |
2-layers GRU | 83.03 | 82.4 | 82.57 |
3-layers CNN | 81.87 | 81.51 | 81.58 |
CNN-1LSTM | 83.86 | 83.55 | 83.62 |
CNN-2LSTM | 83.63 | 83.18 | 83.33 |
CNN-1GRU | 83.42 | 83.02 | 83.13 |
CNN-2GRU | 83.52 | 82.95 | 83.1 |