遇见数据集

Radiomics analysis code and feature matrix for gallstone dual-energy CT virtual monoenergetic image study

收藏
Zenodo2026-10-30 更新2026-11-05 收录
官方服务:

资源简介:

This repository contains the R analysis code and the pre-processed radiomic feature matrix reproducing the primary 40-keV radiomic model reported in the manuscript "Radiomic Separability of Imaging-Defined Gallstone Attenuation Phenotypes Varies Across Dual-Energy CT Virtual Monoenergetic Images" (Frontiers in Medicine). The code performs stratified 70/30 train/test splitting, Z-score scaling, LASSO feature selection with 5-fold cross-validation, radiomic score construction, and held-out AUC evaluation with DeLong 95% confidence intervals and the Youden-optimal threshold. The feature matrix (40kev.xlsx) contains 99 patients and 235 pre-processed radiomic features; the first column is the label (0 = non-negative, 1 = CT-negative).

提供机构:
Zenodo
创建时间:
2026-10-30
二维码
社区交流群
二维码
科研交流群
商业服务