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

Read full text
Authors: Xiya Guan, Wen Zhu, Fangxiang Wu
Cited by: 1

Abstract

There was a mistake in table 2 as published. In the second row ("Baseline", "None"), the Precision (Pre, %) value was incorrectly reported as "87.06". The correct value is "87.16". The corrected table 2 appears below. The original version of this article has been updated.The IoU value in the first sentence of the last paragraph of Section 4.4.3 "Effectiveness of the DCF block and hierarchical configuration" was incorrectly reported as "77.94%" instead of "76.42%".A correction has been made to the section 4.4.3 Effectiveness of the DCF block and hierarchical configuration, last paragraph: "Among all configurations, the equal-weight combination of BCE and Dice loss achieves the best overall performance, yielding a DSC of 85.89% and an IoU of 76.42%. This result indicates that balanced optimization of pixel-level discrimination and overlap consistency is particularly beneficial for pulmonary nodule segmentation, especially in challenging cases involving small nodules and ambiguous boundaries. Therefore, the equal-weight BCE-Dice loss is adopted in the final model."The original version of this article has been updated.

Fields of Study & Subjects

Lung Cancer Diagnosis and Treatment COVID-19 diagnosis using AI Brain Tumor Detection and Classification

Publication details

Journal / source
Frontiers in Oncology
Volume
16
Pages
1905124–1905124
Publication date
9 July 2026
Language
EN