Correction to: Incremental 2D self-labelling for effective 3D medical volume segmentation with minimal annotations
Matthew Anderson, Maged Habib, David H. Steel, Boguslaw Obara
Abstract
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsDigital Image Processing Techniques · Medical Image Segmentation Techniques · 3D Shape Modeling and Analysis
Correction to: Anderson et al. BMC Medical Imaging (2025) 25:450.
10.1186/s12880-025-01991-9.
Following the publication of the original article, the author reported two errors.
- In Fig. 3, the letter “S” in the x‑label appears in a different color in the published version.
- In Fig. 6, the published caption contains “(NSI ≈ 0.3{0.7)”, which is incorrect. The correct caption should be:
“Final test DSC of the 2D self-labelling model when initial training is performed using different single-slice ground-truth annotations from varying positions within the training volumes on the BRAIN_1 dataset. Results show that starting from slices near the centre of the volume (NSI ≈ 0.3–0.7) yields better performance, while peripheral slices lead to significantly worse outcomes.”
The original article has been corrected.