While the decrease in overall discriminative ability of our model on the external test sets did not reach statistical significance as previously reported in brain tumor segmentation by AlBadawy et al [ 11 ] and in PET/MRI enhancement by Chen et al [ 10 ], our findings are in line with reports of slight degradation in CT head and neck squamous cell carcinoma classification by Kann et al [ 9 ] and histopathological polyp classification by Wei et al [ 26 ].
← all excerpts
Application of a validated prostate MRI deep learning system to independent same-vendor multi-institutional data: demonstration of transferability.
1
—
—