Abstract
Following publication of this article [1], concerns were raised about the availability of the underlying dataset and about the methodology used in this study. The authors clarified that the original image database used in this study is available at 10. 6084/m9.figshare.9928529. In the original database (consisting of 260 indirect immunofluorescence images categorized into four patterns), there were similarities and overlapping regions between some images assigned to the training set and the test set, which raised concerns about the suitability of the methodology and the reliability of the conclusions. In order to address these concerns, the authors carried out a supplemental reanalysis of a revised image database with similar and overlapping images removed. This notice reports the results of the supplemental reanalysis.
Cite
CITATION STYLE
Li, J., Tseng, K. K., Hsieh, Z. Y., Yang, C. W., & Huang, H. N. (2020, July 1). Correction: Staining pattern classification of antinuclear autoantibodies based on block segmentation in indirect immunofluorescence images (PLoS ONE (2020) 9:12 (e113132) DOI: 10.1371/journal.pone.0113132). PLoS ONE. Public Library of Science. https://doi.org/10.1371/journal.pone.0236463
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.