RJafroc (version 1.3.2)

DfFroc2Afroc: Convert an FROC dataset to an AFROC dataset

Description

Converts an FROC dataset to a AFROC dataset, where only the highest rated mark on each non-diseased case is counted and all lesion localizations are counted

Usage

DfFroc2Afroc(dataset)

Arguments

dataset

The dataset to be converted, RJafroc-package.

Value

An AFROC dataset

Details

The first list member of the AFROC dataset is NL, whose third dimension has length (K1 + K2), the total number of cases. The ratings of cases (K1 + 1) through (K1 + K2) are -Inf. In an AFROC dataset FPs are only possible on non-diseased cases. The second member of the list is LL. Its third dimension has length K2, the total number of diseased cases. This is because LLs are only possible on diseased cases. The structure is shown below:

• NL Ratings array [1:I, 1:J, 1:(K1+K2), 1:maxNL], of non-lesion localizations, NLs

• LL Ratings array [1:I, 1:J, 1:K2, 1:maxLL], of lesion localizations, LLs

• lesionVector array [1:K2], number of lesions per diseased case

• lesionID array [1:K2, 1:maxLL], labels of lesions on diseased cases

• lesionWeight array [1:K2, 1:maxLL], weights (or clinical importances) of lesions

• dataType "FROC", the data type

• modalityID [1:I] inherited modality labels

• readerID [1:J] inherited reader labels

Examples

# NOT RUN {
afrocDataSet <- DfFroc2Afroc(dataset05)
p <- PlotEmpiricalOperatingCharacteristics(afrocDataSet, trts = 1, rdrs = 1, opChType = "wAFROC")
print(p$Plot)
str(afrocDataSet)

# }