A Cascaded Convolutional Neural Network for Assessing Signal Quality of Dynamic ECG

Table 2

The baseline model configuration.

---

Baseline model

---

Input (1 × 512)

↓

W1_block2_32

↓

Max_pooling_1 × 4

↓

W2_block2_64

↓

Max_pooling_1 × 4

↓

W3_block2_128

↓

Max_pooling_1 × 4

↓

Conv_1 × 1_32

↓

Fc_128

↓

Softmax

---

Vertical arrows indicate sequential connections in the network. Note that W1, W2, and W3 are the number of block2 items. For example, W1_block2_32 denotes that W1 block2 items are connected sequentially, each of which outputs 32 feature maps. Max_pooling_1 × 4 denotes a max pooling with a size of 1 × 4.