RJafroc (version 1.0.1)

SsPowerGivenJK: Statistical power for specified numbers of readers and cases in an ROC study

Description

Calculate the statistical power for specified numbers of readers J, cases K, analysis method and DBM or OR variances components

Usage

SsPowerGivenJK(J, K, effectSize, method, option = "ALL", alpha = 0.05, ...)

Arguments

J

The number of readers in the pivotal study

K

The number of cases in the pivotal study

effectSize

The effect size to be used in the calculation, the sign is unimportant, see Ch 11 in book for guidance.

method

"DBMH" or "ORH"

option

"RRRC", "FRRC", "RRFC" or "ALL"; the default is "ALL"

alpha

The significance level, default is 0.05.

...

Other necessary parameters, OR or DBM variance components, see details

Value

The expected statistical power.

Details

Regarding other parameters (...) needed are either the set of of DBM variance components, i.e, (varYTR, varYTC, and varYEps), or the set of OR covariance matrix elements, the treatment-reader variance and number of cases in pilot study i.e, (cov1, cov2, cov3, varEps, varTR and KStar).

If both of are given, DBM variance components are used and the OR values are ignored.

Either numeric values, for example, of varYTR, varYTC, varYEps can be supplied, provided they are in that order, or the function call must explicitly state, for example, cov1 = value1, cov2 = value2, cov3 = value3, varTR = value4, varEps = value5, KStar = value6, i.e., in any order.

References

Hillis SL, Obuchowski NA, Berbaum KS (2011). Power Estimation for Multireader ROC Methods: An Updated and Unified Approach. Acad Radiol, 18, 129--142.

Hillis SL, Obuchowski NA, Schartz KM, Berbaum KS (2005). A comparison of the Dorfman-Berbaum-Metz and Obuchowski-Rockette methods for receiver operating characteristic (ROC) data. Statistics in Medicine, 24(10), 1579--607.

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 {
## An example of sample size calculation with DBM variance componements
retDbm <- StSignificanceTesting(data = dataset02, 
FOM = "Wilcoxon", method = "DBMH")
effectSize <- retDbm$ciDiffTrtRRRC$Estimate
varCompDBM <- retDbm$varComp
varYTR <- varCompDBM$varComp[3]
varYTC <- varCompDBM$varComp[4]
varYEps <- varCompDBM$varComp[6]
## should give close to 80% power for RRRC
SsPowerGivenJK(6, 251, effectSize, "DBMH", varYTR = varYTR, varYTC = varYTC, 
             varYEps = varYEps)
                     
## An example of sample size calculation with OR variance componements.
retOR <- StSignificanceTesting(data = dataset02, 
FOM = "Wilcoxon", covEstMethod = "Jackknife", method = "ORH")
effectSize <- retOR$ciDiffTrtRRRC$Estimate
varCompOR <- retOR$varComp
varTR <- varCompOR$varCov[2]
cov1 <- varCompOR$varCov[3]
cov2 <- varCompOR$varCov[4]
cov3 <- varCompOR$varCov[5]
varEps <- varCompOR$varCov[6]
KStar <- length(dataset02$NL[1,1,,1])
## same sample size as above, different method, should again give close to 80% power for RRRC
SsPowerGivenJK(6, 251, effectSize, "ORH", cov1 = cov1, cov2 = cov2, cov3 = cov3, 
             varEps = varEps, varTR = varTR, KStar = KStar)

# }