carsaesi/EHRsourceVariability: Measurement and visualization of biomedical data heterogeneity across data sources
The 'EHRsourceVariability' package contains functions to measure and visualize biomedical data heterogeneity across data sources (such as locations, hospitals, professionals, etc.). The multi-source variability measurement methods include two metrics. The first measures the dissimilarity of a data source to a global central tendency of sources, namely the Source Probabilistic Outlyingness (SPO) metric. The second measures the global variability among all the data sources in a repository, namely the Global Probabilistic Deviation (GPD) metric. The metrics are complemented with an exploratory visualization of the variability among data sources, namely the Multi-Source Variability (MSV) plot. These methods serve to highlight anomalous behaviors in the data of specific sources, detect groups of sources with similar data, or provide an indicator of concordance among data sources.
Getting started
Browse package contents
Package details |
|
|---|---|
| Maintainer | Carlos Sáez <carsaesi@upv.es> |
| License | Apache License 2.0 | file LICENSE |
| Version | 0.0.1 |
| Package repository | View on GitHub |
| Installation |
Install the latest version of this package by entering the following in R:
|
