Furthermore, when p-values were adjusted for multiple comparisons using FDR correction, some results that were nominally significant did not remain significant.
← all excerpts
Deep learning-based classification of benign and malignant breast microcalcifications in mammography.
2
—
—
The sentences
Interestingly, paired t-tests suggested marginal significance between B3 and B0/B1 in terms of the weighted F1-score; however, this was not corroborated by the non-parametric Wilcoxon test, potentially due to the non-normal distribution of B3’s F1-scores.