Qipeng Ma, Linbo Xie, Li Peng. Application of Improved Convolutional Neural Network in Medical Image Segmentation[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141020
- Laser & Optoelectronics Progress
- Vol. 57, Issue 14, 141020 (2020)
Fig. 1. Structural diagram of DenseBlock
Fig. 2. Dilated convolutions under different dilation rates when convolution kernel is 3×3. (a) Dilation rate is 1; (b) dilation rate is 2
Fig. 3. Iterative process of CRF
Fig. 4. Structure of CRF-RNN
Fig. 5. DenseU-net-CRFRNN model
Fig. 6. Four modal slice images. (a) T1 modal slice image; (b) T1c modal slice image; (c) T2 modal slice image; (d) Flair modal slice image
Fig. 7. Comparison between results segmented by different models and annotations of experts. (a) Two-dimensional brain tumor slices; (b) annotations of experts; (c) results segmented by DenseU-net model; (d) results segmented by DenseU-net-CRFRNN model
Fig. 8. Segmentation result of DenseU-net model
Fig. 9. Segmentation result of DenseU-net-CRFRNN model
|
Table 1. Accuracy comparison of different methods
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20