Magnification Generalization For Histopathology Image Embedding

  title={Magnification Generalization For Histopathology Image Embedding},
  author={Milad Sikaroudi and Benyamin Ghojogh and Fakhri Karray and Mark Crowley and Hamid R. Tizhoosh},
  journal={2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)},
Histopathology image embedding is an active research area in computer vision. Most of the embedding models exclusively concentrate on a specific magnification level. However, a useful task in histopathology embedding is to train an embedding space regardless of the magnification level. Two main approaches for tackling this goal are domain adaptation and domain generalization, where the target magnification levels may or may not be introduced to the model in training, respectively. Although… 

Figures and Tables from this paper

Magnification Prior: A Self-Supervised Method for Learning Representations on Breast Cancer Histopathological Images
The proposed method, Magnification Prior Contrastive Similarity (MPCS), enables self-supervised learning of representations without labels on small-scale breast cancer dataset BreakHis by exploiting magnification factor, inductive transfer, and reducing human prior.
Hospital-Agnostic Image Representation Learning in Digital Pathology
A domain generalization technique is leveraged in this study to improve the generalization capability of a Deep Neural Network to an unseen histopathology image set (i.e., from an unseen hospital/trial site) in the presence of domain shift.
Generalizing Nucleus Recognition Model in Multi-source Ki67 Immunohistochemistry Stained Images via Domain-Specific Pruning
This paper proposes a novel method to improve domain generalization by searching the domain-agnostic subnetwork in a domain merging scenario and yields excellent performance in multiclass nucleus recognition of Ki67 IHC images, especially in the lost category cases.


Deep learning for magnification independent breast cancer histopathology image classification
Experimental results show that the magnification independent CNN approach improved the performance of magnification specific model, and the results in this limited set of training data are comparable with previous state-of-the-art results obtained by hand-crafted features.
Classifying histopathology whole-slides using fusion of decisions from deep convolutional network on a collection of random multi-views at multi-magnification
A deep convolutional neural network (CNN) based solution is proposed, where images from random number of regions of the tissue section at multiple magnifications are analysed without any necessity of view correspondence across magnifications.
Recognizing Magnification Levels in Microscopic Snapshots
  • Manit Zaveri, S. Kalra, H. Tizhoosh
  • Computer Science
    2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)
  • 2020
This paper extracts deep features of the images available on TCGA dataset with known magnification to train a classifier for magnification recognition, and compared the results with LBP, a well-known handcrafted feature extraction method.
Pan-cancer diagnostic consensus through searching archival histopathology images using artificial intelligence
The key finding of this validation study was that computational consensus appears to be possible for rendering diagnoses if a sufficiently large number of searchable cases are available for each cancer subtype.
Breast cancer histopathological image classification using Convolutional Neural Networks
This method aims to allow using the high-resolution histopathological images from BreaKHis as input to existing CNN, avoiding adaptations of the model that can lead to a more complex and computationally costly architecture.
A Dataset for Breast Cancer Histopathological Image Classification
A dataset of 7909 breast cancer histopathology images acquired on 82 patients, which is now publicly available from, aimed at automated classification of these images in two classes, which would be a valuable computer-aided diagnosis tool for the clinician.
Domain Generalization via Model-Agnostic Learning of Semantic Features
This work investigates the challenging problem of domain generalization, i.e., training a model on multi-domain source data such that it can directly generalize to target domains with unknown statistics, and adopts a model-agnostic learning paradigm with gradient-based meta-train and meta-test procedures to expose the optimization to domain shift.
Fisher Discriminant Triplet and Contrastive Losses for Training Siamese Networks
The FDT and FDC loss functions are designed based on the statistical formulation of the Fisher Discriminant Analysis (FDA), which is a linear subspace learning method and show the effectiveness of the proposed loss functions.
FaceNet: A unified embedding for face recognition and clustering
A system that directly learns a mapping from face images to a compact Euclidean space where distances directly correspond to a measure offace similarity, and achieves state-of-the-art face recognition performance using only 128-bytes perface.
ImageNet classification with deep convolutional neural networks
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.