Data for "Vasculature segmentation in 3D hierarchical phase-contrast tomography images of human kidneys"

This repository contains all the data associated with the Kaggle competition "SenNet + HOA - Hacking the Human Vasculature in 3D", and is part of the paper "Vasculature segmentation in 3D hierarchical phase-contrast tomography images of human kidneys". It is structured as follows: competition-data: Contains the HiP-CT imaging data (test and train sets) and ground truth labels (where available, provided as .tif image files as well as Run-Length Encoded CSV files, see data page on competition website for details). additional-dataset-files: Contains HiP-CT imaging additional dataset used for analysis in the paper, along with predictions from top-5 winning teams and the gold standard labels as Run-Length Encoded CSV files. whole-kidney-test-set: Contains the HiP-CT imaging dataset for the whole kidneys for two test datasets used in the competition (without gold standard labels). The competition dataset contains half kidneys (with gold standard labels). prediction-files-for-teams: Contains predictions for competition dataset and whole-kidney-test-set for top-5 winning solutions, provided as Run-Length Encoded CSV files. source-data-for-figures: Contains source data files for the figures in manuscript. The raw imaging data is also available on the Human Organ Atlas portal (see Supplementary Information in manuscript for DOIs): https://human-organ-atlas.esrf.fr/ The associated manuscript can be found at https://doi.org/10.1101/2024.08.25.609595.