KM-PoPiGo User Guide

Feature Overview

The KM-PoPiGo platform is a versatile tool accessible both as an online web application and as open source code for local deployment. It is structured around a central IPD Reconstruction interface—which integrates Example Cases and Project Management features to streamline the workflow—and a dedicated Benchmark Dataset repository.

1. IPD Reconstruction (Core Interface) Serving as the central interface for digitizing Kaplan-Meier curves, this workspace orchestrates the entire extraction workflow. To support rigorous post-reconstruction validation, it features an integrated R code generator that produces ready-to-use scripts for statistical assessment (e.g., HR estimation) and graphical reproduction.

2. Example Cases (Integrated Feature) Seamlessly integrated into the reconstruction interface, this module offers standardized training sets of KM curve images paired with true survival statistics and comprehensive annotations. These cases enable users to rapidly familiarize themselves with the KM-PoPiGo workflow and validate the precision of the survival statistics derived from their reconstruction results.

3. Project Management (Save & Load) Designed to ensure reproducibility, this feature enables users to archive and restore their workspace locally.

  • Save Project: Encapsulates the comprehensive workspace state (including the KM curve image, axis calibration, and marked data points) into a JSON file for export.
  • Load Project: Restores a previously archived session, allowing for continued editing or retrospective review of the reconstruction process.

4. Benchmark Dataset (Independent Resource) This module serves as an open-access repository hosting three distinct benchmarking datasets: the Real IPD Dataset, the Published KM Dataset, and the Synthetic KM Dataset. These resources are curated to facilitate comprehensive algorithmic validation, enabling users to rigorously evaluate distributional fidelity and the consistency of survival statistics.

Availability & Requirements

  • Web Server: The tool is freely available online at kmpopigo.github.io
  • Browser Compatibility: Optimized for Chrome, Edge, and Firefox.
  • Source Code: Source code are available in the GitHub Repository.