Measuring Domain Shift for Deep Learning in Histopathology

The high capacity of neural networks allows fitting models to data with high precision, but makes generalization to unseen data a challenge. If a domain shift exists, i.e. differences in image statistics between training and test data, care needs to be taken to ensure reliable deployment in real-wor...

Full description

Bibliographic Details
Main Authors:	Eilertsen, G. (Author), Lundstrom, C. (Author), Stacke, K. (Author), Unger, J. (Author)
Format:	Article
Language:	English
Published:	Institute of Electrical and Electronics Engineers Inc. 2021
Subjects:	Article Classification (of information) colon adenocarcinoma colon tissue convolutional neural network Convolutional neural networks deep learning Deep learning Deep Learning Digital pathologies domain shift eosin feature extraction Feature extraction hematoxylin High-precision histopathology human human tissue Humans image classification Image statistics Internal representation learning (artificial intelligence) lymph node lymph node metastasis machine learning Medical imaging Modeling behavior Neural Networks, Computer pathology Pathology quantitative analysis Real-world scenario sensitivity analysis tumor cell tumor classification Tumor classification unsupervised learning unsupervised machine learning Whole slide images
Online Access:	View Fulltext in Publisher

Internet

View Fulltext in Publisher

Measuring Domain Shift for Deep Learning in Histopathology

Internet

Similar Items