Další formáty:
BibTeX
LaTeX
RIS
@article{44568, author = {Hecht, Helge and Mhd Hasan, Sarhan and Popovici, Vlad and Popovici, Vlad and Sarhan, Hasan}, article_location = {Basel}, article_number = {18}, doi = {http://dx.doi.org/10.3390/app10186427}, keywords = {digital pathology; image registration; deep learning; disentangled autoencoder}, language = {eng}, issn = {2076-3417}, journal = {APPLIED SCIENCES}, title = {Disentangled Autoencoder for Cross-Stain Feature Extraction in Pathology Image Analysis}, url = {https://doi.org/10.3390/app10186427}, volume = {10}, year = {2020} }
TY - JOUR ID - 44568 AU - Hecht, Helge - Mhd Hasan, Sarhan - Popovici, Vlad - Popovici, Vlad - Sarhan, Hasan PY - 2020 TI - Disentangled Autoencoder for Cross-Stain Feature Extraction in Pathology Image Analysis JF - APPLIED SCIENCES VL - 10 IS - 18 SP - 1-14 EP - 1-14 PB - MDPI SN - 2076-3417 KW - digital pathology KW - image registration KW - deep learning KW - disentangled autoencoder UR - https://doi.org/10.3390/app10186427 N2 - A novel deep autoencoder architecture is proposed for the analysis of histopathology images. Its purpose is to produce a disentangled latent representation in which the structure and colour information are confined to different subspaces so that stain-independent models may be learned. For this, we introduce two constraints on the representation which are implemented as a classifier and an adversarial discriminator. We show how they can be used for learning a latent representation across haematoxylin-eosin and a number of immune stains. Finally, we demonstrate the utility of the proposed representation in the context of matching image patches for registration applications and for learning a bag of visual words for whole slide image summarization. ER -
HECHT, Helge, Sarhan MHD HASAN a Vlad POPOVICI. Disentangled Autoencoder for Cross-Stain Feature Extraction in Pathology Image Analysis. \textit{APPLIED SCIENCES}. Basel: MDPI, 2020, roč.~10, č.~18, s.~1-14. ISSN~2076-3417. Dostupné z: https://dx.doi.org/10.3390/app10186427.
|