Tumor segmentation in whole-slide images of histology slides is an important step towards computer-assisted diagnosis. In this work, we propose a tumor segmentation framework based on the novel concept of persistent homology profiles (PHPs). For a given image patch, the homology profiles are derived by efficient computation of persistent homology, which is an algebraic tool from homology theory. We propose an efficient way of computing topological persistence of an image, alternative to simplicial homology. The PHPs are devised to distinguish tumor regions from their normal counterparts by modeling the atypical characteristics of tumor nuclei. We propose two variants of our method for tumor segmentation: one that targets speed without compromising accuracy and the other that targets higher accuracy. The fast version is based on selection of exemplar image patches from a convolution neural network (CNN) and patch classification by quantifying the divergence between the PHPs of exemplars and the input image patch. Detailed comparative evaluation shows that the proposed algorithm is significantly faster than competing algorithms while achieving comparable results. The accurate version combines the PHPs and high-level CNN features and employs a multi-stage ensemble strategy for image patch labeling. Experimental results demonstrate that the combination of 1 arXiv:1805.03699v1 [cs.CV] 9 May 2018 PHPs and CNN features outperforms competing algorithms. This study is performed on two independently collected colorectal datasets containing adenoma, adenocarcinoma, signet and healthy cases. Collectively, the accurate tumor segmentation produces the highest average patch-level F1-score, as compared with competing algorithms, on malignant and healthy cases from both the datasets.Overall the proposed framework highlights the utility of persistent homology for histopathology image analysis.
Several reports have demonstrated the use of whole-slide imaging (WSI) for primary pathological diagnosis, but no such studies have been published from Asia. We retrospectively collected 1070 WSI specimens from 900 biopsies and small surgeries conducted in nine hospitals. Nine pathologists, who participated in this study, trained for the College of American Pathologists guidelines, reviewed the specimens and made diagnoses based on digitized, 20Â or 40Â optically magnified images with a WSI scanner. After a washout interval of over 2 weeks, the same observers reviewed conventional glass slides and diagnosed them by light microscopy. Discrepancies between microscopy-and WSI-based diagnoses were evaluated at the individual institutes, and discrepant cases were further reviewed by all pathologists. Nine diagnoses (0.9%) showed major discrepancies with significant clinical differences between the WSI-and microscopy-based diagnoses, and 37 (3.5%) minor discrepancies occurred without a clinical difference. Eight out of nine diagnoses with a major discrepancy were considered concordant with the microscopy-based diagnoses. No association was observed between the level of discrepancy and the organ type, collection method, or digitized optical magnification. Our results indicate the availability of WSI-based primary diagnosis of biopsies and small surgeries in routine daily practice.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.