https://github.com/cns-iu/hra-vccf-2d-kaggle-2023

This repository contains all the code for the paper "Segmentation of human microvasculature in histological images using deep learning for a data-driven Vasculature Common Coordinate Framework".
https://github.com/cns-iu/hra-vccf-2d-kaggle-2023

Last synced: about 1 month ago
JSON representation

This repository contains all the code for the paper "Segmentation of human microvasculature in histological images using deep learning for a data-driven Vasculature Common Coordinate Framework".

• Host: GitHub
• URL: https://github.com/cns-iu/hra-vccf-2d-kaggle-2023
• Owner: cns-iu
• License: mit
• Created: 2023-05-01T18:18:09.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-09-29T17:20:47.000Z (over 1 year ago)
• Last Synced: 2024-02-06T04:42:37.148Z (11 months ago)
• Language: Jupyter Notebook
• Size: 1.34 MB
• Stars: 0
• Watchers: 4
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Segmentation of human microvasculature in histological images using deep learning for a data-driven Vasculature Common Coordinate Framework

The Vascular Common Coordinate Framework (VCCF) uses the human blood vasculature as a roadmap to navigate across the  many scales of the human body to create an address system for all 37 trillion cells. A significant challenge is the lack of literature and experimental data describing the spatial topology of microvasculature structures, such as arterioles, venules, and capillaries. We propose a global challenge to build machine learning algorithms that can detect microvascular structures in human tissue data. This paper presents the design of the “HuBMAP – Hacking the Human Vasculature'' challenge, which tasks participants to develop robust and scalable algorithms to segment microvasculature instances in human kidney histology images. We also present the planned analysis of the winning algorithms. The competition's winning models, strategies, and data analysis will benefit the VCCF efforts and provide potential ways to discover more about microvasculature arrangements throughout the human body that have yet to be examined.