Related Experiment Video
Updated: May 31, 2026

Inverse Probability of Treatment Weighting (Propensity Score) using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Regional medical data mining system
Raul Robu1, Vasile Stoicu-Tivadar
1Politehnica University from Timisoara, Romania. raul.robu@aut.upt.ro
Abstract:
This paper suggests a solution to acquire medical data from hospitals located in a region (addressing especially the DKMT Euroregion), and then perform data mining. The medical data from the hospital databases are exported in XML format, according to HL7 CDA standard. Afterwards, they are automatically centralized on a server in a database using web services calls. The data will be analyzed with the data mining tool WEKA. Data of interest are converted into ARFF format and loaded into WEKA. The next stage consists in preprocessing and analyzing the data based on the algorithms provided by WEKA, having as a goal several relevant medical conclusions. WEKA application interface has been improved to facilitate the process of performing predictions.
