Correction: DPCrossU-Net: a dual-branch parallel CNN–Transformer network for lung nodule segmentation - Report - MDSpire

Correction: DPCrossU-Net: a dual-branch parallel CNN–Transformer network for lung nodule segmentation

  • By

  • Xiya Guan

  • Wen Zhu

  • Fangxiang Wu

  • July 9, 2026

  • 0 min

Share

Clinical Report: Erratum on DPCrossU-Net for Lung Nodule Segmentation

Background

Accurate segmentation of pulmonary nodules in CT scans is essential for early lung cancer detection and diagnosis. The DPCrossU-Net model aims to improve segmentation performance by utilizing a dual-branch architecture that combines CNN and Transformer methodologies.

Data Highlights

MetricOriginal ValueCorrected Value
Precision (Pre, %)87.0687.16
IoU (%)77.9476.42

Key Findings

  • The Precision value for the baseline was corrected from 87.06% to 87.16%.
  • The IoU value was corrected from 77.94% to 76.42%.
  • The equal-weight combination of BCE and Dice loss achieved a DSC of 85.89% and an IoU of 76.42%.

Clinical Implications

The corrections made to the DPCrossU-Net study highlight the importance of precise reporting in clinical research. Accurate metrics are vital for clinicians to evaluate the effectiveness of segmentation models in lung cancer screening.

Conclusion

The erratum clarifies previously reported metrics.

Related Resources & Content

  1. Guan X, Zhu W, Wu F, Frontiers in Oncology, 2026 -- Erratum: DPCrossU-Net - A Dual-Branch Parallel CNN and Transformer Architecture for Segmenting Lung Nodules
  2. npj Digital Medicine — GLANCE: A Novel Approach for Enhanced Nodule Segmentation through Continuous Global-Local Interaction and Consensus Fusion
  3. npj Digital Medicine — Hierarchical Mamba-CNN Transducer for Enhanced Liver Tumor Segmentation in CT Imaging
  4. Utilizing Anatomical Focus to Enhance Segmentation of the Enlarged Pancreatic Duct in Abdominal CT Imaging
  5. Final Recommendation Statement: Lung Cancer: Screening | United States Preventive Services Taskforce
  6. Guidelines for Management of Incidental Pulmonary Nodules Detected on CT Images: From the Fleischner Society 2017 | Radiology
  7. Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening | New England Journal of Medicine

Original Source(s)

Correction: DPCrossU-Net: a dual-branch parallel CNN–Transformer network for lung nodule segmentation

  • by Xiya Guan, Wen Zhu, Fangxiang Wu

  • July 9, 2026

Related Content

The expression landscape and heterogeneity of nectin-4, HER2, and Trop-2 in urothelial carcinoma: clinicopathological correlations and implications for ADC-guided precision therapy

Systemic immune remodeling following curative (R0) resection of colorectal liver metastases

Regression of retinal capillary hemangioblastoma with systemic belzutifan in von Hippel–Lindau disease: a case report