RJafroc (version 1.0.1)

SsFROCPowerGivenJK: Statistical power in ROC and FROC paradigms from an ROC/FROC/LROC NH binned dataset

Description

Compares statistical power using ROC and FROC paradigms, over a range of ROC effect sizes, from variability information obtained from a null hypothesis binned dataset, which can be in ROC/FROC/LROC paradigms, for J readers and K cases in the pivotal study.

Usage

SsFROCPowerGivenJK(dataset, trts, rdrs, effectSizeROC, J, K)

Arguments

dataset

The pilot dataset to be analyzed, see RJafroc-package, for variability information. The dataType can be "ROC", "FROC", or "LROC".

trts

The indices of the modalities in the pilot dataset that will be regarded as representative of null hypothesis modalities. Two or more modalities, specified by indices, e.g., c(1,2,3).

rdrs

The indices of the readers in the pilot dataset that will be regarded as representative of the NH readers; this can be used for example to exclude an atypical reader.

effectSizeROC

Array, the range of expected ROC effect sizes to scan; see book Chapter 11 for guidelines, e.g., seq(0.01, 0.09, 0.005).

J

The number of readers in the pivotal study.

K

The number of cases in the pivotal study.

Value

The returned list contains following items.

powerROC

Array, length(effectSizeROC), the statistical power using ROC methodology.

powerwAFROC

Array, length(effectSizeROC), the statistical power using wAFROC methodology.

## potential project for summer student

References

Chakraborty DP (2017) Observer Performance Methods for Diagnostic Imaging - Foundations, Modeling, and Applications with R-Based Examples, CRC Press, Boca Raton, FL. https://www.crcpress.com/Observer-Performance-Methods-for-Diagnostic-Imaging-Foundations-Modeling/Chakraborty/p/book/9781482214840

Examples

# NOT RUN {
SsFROCPowerGivenJK(dataset04, trts = c(1,2), rdrs = c(1,2,3,4), 
   effectSizeROC = seq(0.01, 0.09, 0.005), J = 5, K = 200)

##SsFROCPowerGivenJK(datasetCadLroc, trts = 1, rdrs = seq(2,9), 
##    effectSizeROC = seq(0.01, 0.09, 0.005), J = 5, K = 200)
   
# }
# NOT RUN {
# }