Hepatocellular carcinoma (HCC) survival prediction and clinical factor ranking through computational intelligence
HCC Survival dataset on the University of California Irvine Machine Learning Repository: https://archive.ics.uci.edu/ml/datasets/HCC+Survival
Software: R platform with packages: easypackages
, randomForest
, caret
, mltools
, MLmetrics
Operating systems: any Linux, any Microsoft Windows, any Mac operating system running R.
To analyze the dataset for survival prediction and for feature ranking, clone/download this GitHub repository, and visit the CODE_SUBMISSION
folder.
For the survival prediction, type on the shell termal console:
Rscript binary_classification_v1a.r
For the feature ranking, type on the shell termal console:
Rscript feature_ranking_v1b.r
The program will print the results into files that you will find in the RIS
folder.
Additional information about this project is available in the following peer-reviewed published article:
Davide Chicco, Luca Oneto. "Computational intelligence identifies alkaline phosphatase (ALP), alpha-fetoprotein (AFP), and hemoglobin levels as most predictive survival factors for hepatocellular carcinoma". Health Informatics Journal 27(1), pages 1-26, 2021.
For any information, please contact Davide Chicco at davidechicco(AT)davidechicco.it or Luca Oneto at luca.oneto(AT)gmail.com