Article
Boundary-aware glomerulus segmentation: toward one-to-many stain generalization
Registro en:
SILVA, Jefferson et al. Boundary-aware glomerulus segmentation: toward one-to-many stain generalization. Computerized Medical Imaging and Graphics, v. 100, p. 1-13, 2022.
1879-0771
10.1016/j.compmedimag.2022.102104
Autor
Silva, Jefferson
Souza, Luiz
Chagas, Paulo
Calumby, Rodrigo
Souza, Bianca
Pontes, Izabelle
Duarte, Angelo
Pinheiro, Nathanael
Santos, Washington
Oliveira, Luciano
Resumen
Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB).
Inova FIOCRUZ.
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). The growing availability of scanned whole-slide images (WSIs) has allowed nephropathology to open new possibilities for medical decision-making over high-resolution images. Diagnosis of renal WSIs includes locating and identifying specific structures in the tissue. Considering the glomerulus as one of the first structures analyzed by pathologists, we propose here a novel convolutional neural network for glomerulus segmentation. Our end-to-end network, named DS-FNet, combines the strengths of semantic segmentation and semantic boundary detection networks via an attention-aware mechanism. Although we trained the proposed network on periodic acid-Schiff (PAS)-stained WSIs, we found that our network was capable to segment glomeruli on WSIs stained with different techniques, such as periodic acid-methenamine silver (PAMS), hematoxylin-eosin (HE), and Masson trichrome (TRI). To assess the performance of the proposed method, we used three public data sets: HuBMAP (available in a Kaggle competition), a subset of the NEPTUNE data set, and a novel challenging data set, called WSI_Fiocruz. We compared the DS-FNet with six other deep learning networks: original UNet, our attention version of U-Net called AU-Net, U-Net++, U-Net3Plus, ResU-Net, and DeepLabV3+. Results showed that DS-FNet achieved equivalent or superior results on all data sets: On the HuBMAP data set, it reached a dice score (DSC) of 95.05%, very close to the first place (95.15%); on the NEPTUNE and WSI_Fiocruz data sets, DS-FNet obtained the highest average DSC, whether on PAS-stained images or images stained with
other techniques. To the best we know, this is the first work to show consistently high performance in a one-to-many-stain glomerulus segmentation following a thorough protocol on data sets from different medical labs.