We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CV

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: Cells are Actors: Social Network Analysis with Classical ML for SOTA Histology Image Classification

Abstract: Digitization of histology images and the advent of new computational methods, like deep learning, have helped the automatic grading of colorectal adenocarcinoma cancer (CRA). Present automated CRA grading methods, however, usually use tiny image patches and thus fail to integrate the entire tissue micro-architecture for grading purposes. To tackle these challenges, we propose to use a statistical network analysis method to describe the complex structure of the tissue micro-environment by modelling nuclei and their connections as a network. We show that by analyzing only the interactions between the cells in a network, we can extract highly discriminative statistical features for CRA grading. Unlike other deep learning or convolutional graph-based approaches, our method is highly scalable (can be used for cell networks consist of millions of nodes), completely explainable, and computationally inexpensive. We create cell networks on a broad CRC histology image dataset, experiment with our method, and report state-of-the-art performance for the prediction of three-class CRA grading.
Subjects: Computer Vision and Pattern Recognition (cs.CV); Social and Information Networks (cs.SI)
Cite as: arXiv:2106.15299 [cs.CV]
  (or arXiv:2106.15299v2 [cs.CV] for this version)

Submission history

From: Mostafa Jahanifar [view email]
[v1] Tue, 29 Jun 2021 12:22:10 GMT (2486kb)
[v2] Sat, 10 Jul 2021 11:57:37 GMT (7174kb)

Link back to: arXiv, form interface, contact.