threerocs {rocbc}

Provides visual comparison of three ROC estimation methods (one-marker version)

Description

This function provides a visual comparison of the Empirical ROC, the Box-Cox ROC, and the Metz binormal semi-parametric estimator of the ROC curve. It also computes the AUC for the curve corresponding to each method.

Usage

threerocs(marker, D, plots)

Arguments

marker

A vector of length n that contains the biomarker scores of all individuals.

D

A vector of length n that contains the true disease status of an individual. It is a binary vector containing 0 for the healthy/control individuals and 1 for the diseased individuals.

plots

Valid inputs are "on" and "off". When set to "on", the user gets a single plot containing the estimated ROC curves using the Empirical, Box-Cox, and Metz methods for each of the two provided markers.

Value

AUC_Empirical

The AUC of the empirical ROC curve.

AUC_Metz

The AUC of the Metz binormal curve (as calculated by MRMCaov package using the "binormal" option).

AUC_BoxCox

The AUC of the Box-Cox based ROC curve.

Author(s)

Leonidas Bantis

References

Box GEP, Cox DR. (1964). An Analysis of Transformations. Journal of the Royal Statistical Society. 26(2):211-252. https://www.jstor.org/stable/2984418

Smith BJ, Hillis SL, Pesce LL (2023). MCMCaov: Multi-Reader Multi-Case Analysis of Variance. R package version 0.3.0, https://github.com/brian-j-smith/MRMCaov.

Smith BJ, Hillis SL (2020). “Multi-reader multi-case analysis of variance software for diagnostic performance comparison of imaging modalities.” In Samuelson F, Taylor-Phillips S (eds.), Proceedings of SPIE 11316, Medical Imaging 2020: Image Perception, Observer Performance, and Technology Assessment, 113160K. doi:10.1117/12.2549075, https://pubmed.ncbi.nlm.nih.gov/32351258.

Examples

set.seed(123)
x <- rgamma(100, shape=2, rate = 8) # generates biomarker data from a gamma
                                 # distribution for the healthy group.
y <- rgamma(100, shape=2, rate = 4) # generates biomarker data from a gamma
                                 # distribution for the diseased group.
scores <- c(x,y)
D=c(pracma::zeros(1,100), pracma::ones(1,100))
out=threerocs(marker=scores, D, plots="on")
summary(out)