Data augmentation based on spatial deformations for histopathology: An evaluation in the context of glomeruli segmentation.

Date publication

juin 2022

Journal

Computer methods and programs in biomedicine

Auteurs

Membres identifiés du Cancéropôle Est :
Pr WEMMERT Cédric

Tous les auteurs :
Allender F, Allègre R, Wemmert C, Dischler JM

Lien Pubmed

https://www.ncbi.nlm.nih.gov/pubmed/35701252

Résumé

The effective application of deep learning to digital histopathology is hampered by the shortage of high-quality annotated images. In this paper we focus on the supervised segmentation of glomerular structures in patches of whole slide images of renal histopathological slides. Considering a U-Net model employed for segmentation, our goal is to evaluate the impact of augmenting training data with random spatial deformations.

Mots clés

Data augmentation, Glomeruli segmentation, Histopathological images, Random spatial deformations

Référence

Comput Methods Programs Biomed. 2022 06;221:106919