AVM clinical data for building prediction models

Published: 4 June 2020| Version 2 | DOI: 10.17632/k263py4fzb.2
Fenghua Chen


The dataset in the file "combined1.csv" included 383 patient's clinical data, including 7 variables: age, sex, location, aneurysm, nidus size, draining type, the number of draining veins, and hemorrhage. Female was defined as 0, male as 1. Superficial AVM was defined as 0, deep AVM as 1, infratentorial AVM as 2. Only superficial draining vein was defined as 0, mixed superficial and deep draining vein as 1. A single draining vein was defined as 1, multiple draining veins as 2. Ruptured AVM was defined as 1, unruptured AVM as 0. The file "randomtestpredict.R" was the source file in RStudio, which was used to build and test prediction models based on the above data file.



Xiangya Hospital Central South University


Arteriovenous Malformation