Journal of Pathology Informatics Journal of Pathology Informatics
Contact us | Home | Login   |  Users Online: 7556  Print this pageEmail this pageSmall font sizeDefault font sizeIncrease font size 


RESEARCH ARTICLE
Year : 2020  |  Volume : 11  |  Issue : 1  |  Page : 38

Constellation loss: Improving the efficiency of deep metric learning loss functions for the optimal embedding of histopathological images


1 TECNALIA, Basque Research and Technology Alliance (BRTA), Parque Tecnológico de Bizkaia, C/ Geldo. Edificio 700, E-48160 Derio - Bizkaia (Spain), Spain
2 TECNALIA, Basque Research and Technology Alliance (BRTA), Parque Tecnológico de Bizkaia, C/ Geldo. Edificio 700, E-48160 Derio; University of the Basque Country, Alameda de Urquijo s/n, 48013 Bilbao, Bizkaia (Spain), Spain

Correspondence Address:
Dr. Artzai Picon
TECNALIA, Basque Research and Technology Alliance (BRTA), Parque Tecnológico de Bizkaia, C/ Geldo. Edificio 700, E-48160 Derio - Bizkaia
Spain
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/jpi.jpi_41_20

Rights and Permissions

Background: Deep learning diagnostic algorithms are proving comparable results with human experts in a wide variety of tasks, and they still require a huge amount of well-annotated data for training, which is often non affordable. Metric learning techniques have allowed a reduction in the required annotated data allowing few-shot learning over deep learning architectures. Aims and Objectives: In this work, we analyze the state-of-the-art loss functions such as triplet loss, contrastive loss, and multi-class N-pair loss for the visual embedding extraction of hematoxylin and eosin (H&E) microscopy images and we propose a novel constellation loss function that takes advantage of the visual distances of the embeddings of the negative samples and thus, performing a regularization that increases the quality of the extracted embeddings. Materials and Methods: To this end, we employed the public H&E imaging dataset from the University Medical Center Mannheim (Germany) that contains tissue samples from low-grade and high-grade primary tumors of digitalized colorectal cancer tissue slides. These samples are divided into eight different textures (1. tumour epithelium, 2. simple stroma, 3. complex stroma, 4. immune cells, 5. debris and mucus, 6. mucosal glands, 7. adipose tissue and 8. background,). The dataset was divided randomly into train and test splits and the training split was used to train a classifier to distinguish among the different textures with just 20 training images. The process was repeated 10 times for each loss function. Performance was compared both for cluster compactness and for classification accuracy on separating the aforementioned textures. Results: Our results show that the proposed loss function outperforms the other methods by obtaining more compact clusters (Davis-Boulding: 1.41 ± 0.08, Silhouette: 0.37 ± 0.02) and better classification capabilities (accuracy: 85.0 ± 0.6) over H and E microscopy images. We demonstrate that the proposed constellation loss can be successfully used in the medical domain in situations of data scarcity.


[FULL TEXT] [PDF]*
Print this article     Email this article
 Next article
 Previous article
 Table of Contents

 Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
 Citation Manager
 Access Statistics
 Reader Comments
 Email Alert *
 Add to My List *
 * Requires registration (Free)
 

 Article Access Statistics
    Viewed2708    
    Printed22    
    Emailed0    
    PDF Downloaded357    
    Comments [Add]    
    Cited by others 4    

Recommend this journal