Correction: DPCrossU-Net: a dual-branch parallel CNN–Transformer network for lung nodule segmentation
-
By
-
Xiya Guan
-
Wen Zhu
-
Fangxiang Wu
-
July 9, 2026
-
Objective:
To correct errors in the published article regarding the DPCrossU-Net architecture for lung nodule segmentation.
Approach:
- Correction of Table 2: The Precision (Pre, %) value for 'Baseline' was corrected from '87.06' to '87.16'.
- Correction of IoU Value: The IoU value was corrected from '77.94%' to '76.42%' in Section 4.4.3.
- Updated Performance Metrics: The equal-weight combination of BCE and Dice loss was confirmed to yield a DSC of 85.89% and an IoU of 76.42%.
Key Findings:
- Precision value corrected from 87.06 to 87.16.
- IoU value corrected from 77.94% to 76.42%.
- Equal-weight BCE–Dice loss provides optimal performance for pulmonary nodule segmentation.
Interpretation:
Conclusion:
The original article has been updated to reflect these corrections.
Sources:
