Research Article
Local Binary Convolutional Neural Networks' Long Short-Term Memory Model for Human Embryos' Anomaly Detection
Table 9
Diagnosis of embryo anomaly in different periods based on accuracy.
| | Model | Time | | 24 h | 36 h | 48 h | 60 h | 72 h |
| | VGG16 + LSTM | 30.4 | 42.8 | 69.2 | 70.1 | 78.9 | | LeNet 5 + LSTM | 22.8 | 30.8 | 52.1 | 55.6 | 55.6 | | Resnet-50 + LSTM | 31.9 | 48.7 | 69.3 | 67.1 | 80.3 | | Inception V3 + LSTM | 30.9 | 41.9 | 67.2 | 66.9 | 79.7 | | MobileNet V2 + LSTM | 29.3 | 35.7 | 59.1 | 65.1 | 65 | | Xception + LSTM | 30.1 | 47.6 | 61.3 | 66.9 | 73.9 | | Baseline CNN + LSTM | 22.8 | 36.8 | 60.2 | 62.3 | 70.1 | | LBCNN + LSTM | 30.4 | 47.9 | 69.8 | 75.2 | 82.8 |
|
|
The proposed approach (LBCNN+LSTM) experimental results declared as bold values.
|
