Acta Photonica Sinica, Volume. 53, Issue 1, 0111003(2024)
Boundary Perception Network for Pathological Image Segmentation
Figures & Tables(12)
Fig. 3. The structure of adaptive shuffle channel attention module
Fig. 6. The segmentation results of different algorithms on GlaS datasets
Fig. 7. The segmentation results of ablation experiments on GlaS datasets
Fig. 8. The segmentation results of different algorithms on MoNuSeg datasets
Table 1. The experimental results with different methods on GlaS datasets（Mean ± Std）
View tableView in ArticleTable 1. The experimental results with different methods on GlaS datasets（Mean ± Std）
Algorithm Dice/% IoU/% ACC/% PRE/% Parameters/（×106） U-Net（2015） 83.40±3.26 71.84±5.86 83.04±3.83 79.99±6.51 31.0 UNet++（2018） 85.11±0.73 74.10±1.10 85.48±0.41 85.83±0.97 39.4 AttentionUNet（2018） 86.30±2.08 75.99±4.00 86.39±2.94 83.99±5.82 63.1 MultiResUNet（2020） 87.20±0.99 77.38±1.55 86.98±0.69 86.29±0.85 59.1 MedT（2021） 82.86±0.97 70.76±1.41 82.46±0.78 83.93±0.99 7.0 TransUNet（2021） 88.67±0.65 79.66±1.06 88.46±0.69 90.00±0.78 421.2 UCTransNet（2022） 89.39±0.68 80.83±1.11 87.93±0.93 87.00±1.65 65.5 BPNet 92.21±0.19 85.55±0.33 92.14±0.15 92.07±0.93 64.9
Table 2. The ablation experimental results on GlaS datasets（Mean ± Std）
View tableView in ArticleTable 2. The ablation experimental results on GlaS datasets（Mean ± Std）
Algorithm Dice/% IoU/% ACC/% PRE/% Parameters/（×106） Baseline 87.09±1.97 78.10±3.33 88.65±2.16 88.54±3.10 63.3 Baseline+BPM 91.79±0.09 84.81±0.16 91.80±0.07 92.87±0.30 64.8 Baseline+BPM+ASCAM 92.21±0.19 85.55±0.33 92.14±0.15 92.07±0.93 64.9
Table 3. The experimental results with different methods on MoNuSeg datasets（Mean ± Std）
View tableView in ArticleTable 3. The experimental results with different methods on MoNuSeg datasets（Mean ± Std）
Algorithm Dice/% IoU/% ACC/% PRE/% Parameters/（×106） U-Net（2015） 77.17±2.84 62.90±3.79 89.90±1.62 65.01±3.68 31.0 UNet++（2018） 78.90±0.57 65.16±0.78 90.98±0.33 67.42±1.32 39.4 AttentionUNet（2018） 75.00±1.86 60.03±2.38 88.40±1.42 60.64±3.39 63.1 MultiResUNet（2020） 79.70±0.61 66.26±0.84 91.52±0.33 69.19±1.61 59.1 MedT（2021） 74.48±0.61 59.34±0.77 89.15±0.16 63.61±3.30 7.0 TransUNet（2021） 80.50±0.39 67.37±0.54 91.40±0.24 73.42±1.48 421.2 UCTransNet（2022） 79.08±0.79 65.58±1.00 91.02±0.53 70.38±1.13 65.5 BPNet 81.18±0.44 68.34±0.65 92.50±0.24 75.46±2.15 64.9
Table 4. The ablation experimental results on MoNuSeg datasets（Mean ± Std）
View tableView in ArticleTable 4. The ablation experimental results on MoNuSeg datasets（Mean ± Std）
Algorithm Dice/% IoU/% ACC/% PRE/% Parameters/（×106） Baseline 77.74±0.73 63.59±0.98 91.47±0.06 73.38±2.74 63.3 Baseline+BPM 79.96±0.46 66.61±0.64 92.02±0.36 73.97±4.21 64.8 Baseline+BPM+ASCAM 81.18±0.44 68.34±0.65 92.50±0.24 75.46±2.15 64.9
Tools
Get Citation
Copy Citation Text
Hong HUANG, Yichuan YANG, Long WANG, Fujian ZHENG, Jian WU. Boundary Perception Network for Pathological Image Segmentation[J]. Acta Photonica Sinica, 2024, 53(1): 0111003
Paper Information
Category:
Received: Jul. 4, 2023
Accepted: Aug. 1, 2023
Published Online: Feb. 1, 2024
The Author Email: HUANG Hong (hhuang@cqu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20