18 June 2020 : Clinical Research
Ensemble Deep Learning Model for Multicenter Classification of Thyroid Nodules on Ultrasound Images
Xi Wei1ACDEG*, Ming Gao2BEF, Ruiguo Yu3BCD, Zhiqiang Liu3BCE, Qing Gu4BC, Xun Liu5BC, Zhiming Zheng6BC, Xiangqian Zheng2BC, Jialin Zhu1ABCDEF, Sheng Zhang1BDDOI: 10.12659/MSM.926096
Med Sci Monit 2020; 26:e926096
Supplementary Table 2 ROI extraction algorithm structure.
| Processing | Layer | Output size | Activation |
|---|---|---|---|
| Down-sampling | conv1_1 | 224×224 | Relu |
| conv1_2 | 224×224 | Relu | |
| pool1 | 112×112 | ||
| conv2_1 | 112×112 | Relu | |
| conv2_2 | 112×112 | Relu | |
| pool2 | 56×56 | ||
| conv3_1 | 56×56 | Relu | |
| conv3_2 | 56×56 | Relu | |
| conv3_3 | 56×56 | Relu | |
| conv3_4 | 56×56 | Relu | |
| pool3 | 28×28 | ||
| conv4_1 | 28×28 | Relu | |
| conv4_2 | 28×28 | Relu | |
| conv4_3 | 28×28 | Relu | |
| conv4_4 | 28×28 | Relu | |
| pool4 | 14×14 | ||
| conv5_1 | 14×14 | Relu | |
| conv5_2 | 14×14 | Relu | |
| Down-sampling [continued] | conv5_3 | 14×14 | Relu |
| conv5_4 | 14×14 | Relu | |
| pool5 | 7×7 | ||
| conv6 | 7×7 | Relu | |
| conv7 | 7×7 | Relu | |
| conv8 | 7×7 | Relu | |
| Up-sampling | deconv1 | 14×14 | |
| deconv2 | 28×28 | ||
| deconv3 | 224×224 |