SurvivalPath:A R package for conducting personalized survival path mapping based on time-series survival data

The survival path mapping approach has been proposed for dynamic prognostication of cancer patients using time-series survival data. The SurvivalPath R package was developed to facilitate building personalized survival path models. The package contains functions to convert time-series data into time-slices data by fixed interval based on time information of input medical records. After the pre-processing of data, under a user-defined parameters on covariates, significance level, minimum bifurcation sample size and number of time slices for analysis, survival paths can be computed using the main function, which can be visualized as a tree diagram, with important parameters annotated. The package also includes function for analyzing the connections between exposure/treatment and node transitions, and function for screening patient subgroup with specific features, which can be used for further exploration analysis. In this study, we demonstrate the application of this package in a large dataset of patients with hepatocellular carcinoma, which is embedded in the package.

Automated summary

A new survival path mapping approach is proposed for dynamic prognostication of cancer patients using time-series survival data. The SurvivalPath R package facilitates the building of personalized survival path models by converting time-series data into time-slices data and visualizing it as a tree diagram. The package also includes functions for analyzing connections between exposure/treatment and node transitions, as well as screening patient subgroups with specific features.