/RIVD

A new and ambitious project

Primary LanguageROtherNOASSERTION

RIVD

There’s R/Pharma and R/medicine, now, let’s get R/IVD started :sunglasses:

The goal of RIVD is to deliver simple tools and adapted outputs for in vitro diagnostic (IVD) assays performances assessment. The results can be used in early development phase, or hopefully for building a validation report in compliance with regulatory frameworks. The package is intended to follow the recommendations formulated in the CLSI guidelines.

Installation

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("Felixmil/RIVD")

Examples

Performance table for assay’s precision.

library(RIVD)
library(knitr)

data(Glucose,package="VCA")

kable(perfPrecision(Glucose, 'result','run', 'day'))
DF %Total SD CV[%]
Withing laboratory 64.77732 100.00000 3.596325 1.4726965
within-day 19.00000 15.14319 1.399483 0.5730889
within-run 20.00000 23.77537 1.753568 0.7180867
error 40.00000 61.08144 2.810694 1.1509803

Visualisation of data generated by the precision assessment protocol

library(RIVD)

data(Glucose,package="VCA")
plotPrecision(Glucose, 'result','run', 'day')