yadevi / liveboost

LiveBoost: Clinical prediction of HBV and HCV related hepatic fibrosis

Please cite our paper:

Wei, Runmin, et al. "Clinical prediction of HBV and HCV related hepatic fibrosis using machine learning." EBioMedicine (2018).

https://www.sciencedirect.com/science/article/pii/S2352396418302810

For sending comments, suggestions, bug reports of LiveBoost, please email to Runmin Wei (RWei AT cc DOT hawaii DOT edu).

Introduction

Welcome to the LiveBoost repository of datasets and code sharing. The goal of this repository is to make our method more transparent and reproducible.

Webtool

https://metabolomics.cc.hawaii.edu/software/LiveBoost/

Datasets

Discovery_set.csv is the discovery dataset contains 490 HBV infected subjects with their histopathologic staging of hepatic fibrosis, Age, AST, ALT, PLT, and FIB4.

Validation_set1.csv is the validation dataset-1 contains 86 HBV infected subjects with their histopathologic staging of hepatic fibrosis, Age, AST, ALT, PLT, and FIB4.

For validation dataset-2, please refer to the original paper: https://doi.org/10.1371/journal.pone.0190455.

For validation dataset-3, please refer to the original paper: https://doi.org/10.1371/journal.pone.0133515.

Code

Code Sharing.Rmd and Code Sharing.pdf provides session information, package requirements, and all ready-to-use codes written in R.

Other files

DT_tune_S02_S34.jpg, DT_tune_S03_S4.jpg, RF_tune_S02_S34.jpg, RF_tune_S03_S4.jpg, GB_tune_S02_S34.jpg, and GB_tune_S03_S4.jpg are about how we optimize hyper-parameter settings for different models.