Superpixel-Based Conditional Random Fields (SuperCRF): Incorporating Global and Local Context for Enhanced Deep Learning in Melanoma Histopathology.

Publication date: Jun 11, 2019

Computational pathology-based cell classification algorithms are revolutionizing the study of the tumor microenvironment and can provide novel predictive/prognosis biomarkers crucial for the delivery of precision oncology. Current algorithms used on hematoxylin and eosin slides are based on individual cell nuclei morphology with limited local context features. Here, we propose a novel multi-resolution hierarchical framework (SuperCRF) inspired by the way pathologists perceive regional tissue architecture to improve cell classification and demonstrate its clinical applications. We develop SuperCRF by training a state-of-art deep learning spatially constrained- convolution neural network (SC-CNN) to detect and classify cells from 105 high-resolution (20x) H&E-stained slides of The Cancer Genome Atlas melanoma dataset and subsequently, a conditional random field (CRF) by combining cellular neighborhood with tumor regional classification from lower resolution images (5, 1.25x) given by a superpixel-based machine learning framework. SuperCRF led to an 11.85% overall improvement in the accuracy of the state-of-art deep learning SC-CNN cell classifier. Consistent with a stroma-mediated immune suppressive microenvironment, SuperCRF demonstrated that (i) a high ratio of lymphocytes to all lymphocytes within the stromal compartment (p = 0.026) and (ii) a high ratio of stromal cells to all cells (p

Open Access PDF

Zormpas-Petridis, K., Failmezger, H., Raza, S.E.A., , Roxanis, Jamin, Y., and Yuan, Y. Superpixel-Based Conditional Random Fields (SuperCRF): Incorporating Global and Local Context for Enhanced Deep Learning in Melanoma Histopathology. 25273. 2019 Front Oncol (9):

Concepts Keywords
Biomarkers Statistical classification
Classifier Precision medicine
CNN Branches of biology
Conditional Tumor microenvironment
Convolution RTT
Deep Learning Melanoma
Eosin Cancer
Hematoxylin Machine learning
Histopathology Medicine
Lymphocytes Health
Melanoma Classification algorithms
Microenvironment Clinical applications
Morphology Classification algorithms
Neural Network Neural network
Nuclei Digital pathology
Oncology
Pathology
Prognosis
Stroma
Stromal
Tumor

Semantics

Type Source Name
disease MESH Melanoma
pathway KEGG Melanoma
disease MESH Histopathology
disease MESH tumor
disease MESH multi
drug DRUGBANK Corticorelin
disease MESH neighborhood

Similar

Original Article

Leave a Comment

Your email address will not be published. Required fields are marked *