Kokiri is a visual analytics approach to compare and characterize cohorts.
Users can interactively compare cohorts by their high dimensional data, explore the driving differences, and characterize the homogeneity and outliers of the cohorts
Kokiri can be utilized to compare any type of cohort. Our focus, however, is on the analysis of genomic data from cancer patients.
🚀 You can try Kokiri yourself at: https://kokiri.jku-vds-lab.at/
Learn more about Kokiri by reading the paper.
For a quick overview of Kokiri, see our preview video:
Your comments and feedback are welcome. Write an email to [email protected] and let us know what you think!
If you have discovered an issue or have a feature suggestion, feel free to create an issue on GitHub.
Klaus Eckelt, Patrick Adelberger, Markus J. Bauer, Thomas Zichner, Marc Streit
Kokiri: Random Forest-Based Cohort Comparison and Characterization
bioRxiv, 2022.
@article{2022_kokiri,
title = {Kokiri: Random-Forest-Based Comparison and Characterization of Cohorts},
author = {Klaus Eckelt and Patrick Adelberger and Markus J. Bauer and Thomas Zichner and Marc Streit},
journal = {IEEE VIS Workshop on Visualization in Biomedical AI},
doi = {10.1101/2022.08.16.503622},
url = {https://www.biorxiv.org/content/10.1101/2022.08.16.503622},
year = {2022}
}