Dataset and Processing Code Accompanying the Paper: "Sensitivity Analysis for Exploring the Variability and Parameter Landscape in Virtual Patient Cohorts of Multi-Vessel Coronary Artery Disease" by Hilhorst et al.

This dataset includes both the synthetic data and the scripts used to process and visualize this data, as described in the paper Sensitivity Analysis for Exploring the Variability and Parameter Landscape in Virtual Patient Cohorts of Multi-Vessel Coronary Artery Disease by Hilhorst et al. This repository contains multiple folders and files related to virtual patient cohort (VPC) generation, surrogate model training, and sensitivity analysis. Below is a brief description of each folder and its contents: Data_preparationThis includes all the data and scripts necessary to prepare datasets for surrogate model training and to generate a VPC.Codes for deriving the correlation structures used in the sensitivity analysis are also provided. Sensitivity_analysisProvides sensitivity indices for the model parameters.Includes code/scripts for generating plots of the sensitivity analysis results and the convergence tests. Surrogate_TrainingContains training data, test data, and the trained surrogate models themselves.It includes scripts to visualize the surrogate model performance (e.g., Bland-Altman plots and histograms). VPC_GenerationScripts to plot and evaluate the synthetic VPC after various filtering steps.Contains scripts to visualize pressure and flow waveforms, create violin plots, and perform the Kolmogorov-Smirnov test to compare distributions. Additional NotesThere is a notation difference in patient IDs:Patient 28 in the data is referred to in the paper as Patient-2L.Patient 604 in the data is referred to in the paper as Patient-3L.Data from the FAME study is not included in this dataset. To access the FAME data, please contact Prof. Dr. P.A.L. Tonino ([email protected]).Code (e.g., for surrogate training, sensitivity analysis, or the 1D Pulse Wave Propagation model) can be obtained by reaching out to Prof. Dr. W. Huberts ([email protected]).