StSignificanceTestingCadVsRad: Significance testing: standalone CAD vs. radiologists

Description

Comparing standalone CAD vs. at least two radiologists interpreting the same cases; standalone CAD means that all the designer-level mark-rating pairs generated by the CAD algorithm are available to the analyst, not just the one or two marks per case displayed to the radiologist (the latter are marks whose ratings exceed a pre-selected threshold). At the very minimum, location-level information, such as in the LROC paradigm, should be used. Ideally, the FROC paradigm should be used. A severe statistical power penalty is paid if one uses the ROC paradigm. See Standalone CAD vs Radiologists chapter, available via download link at site https://github.com/dpc10ster/RJafrocBook/blob/gh-pages/RJafrocBook.pdf

Usage

StSignificanceTestingCadVsRad(
  dataset,
  FOM,
  FPFValue = 0.2,
  method = "1T-RRRC",
  alpha = 0.05,
  plots = FALSE
)

Value

If method = "1T-RRRC" the return value is a list with the following elements:

fomCAD: The observed FOM for CAD.
fomRAD: The observed FOM array for the readers.
avgRadFom: The average FOM of the readers.
avgDiffFom: The mean of the difference FOM, RAD - CAD.
ciAvgDiffFom: The 95-percent CI of the average difference, RAD - CAD.
varR: The variance of the radiologists.
varError: The variance of the error term in the single-treatment multiple-reader OR model.
cov2: The covariance of the error term.
tstat: The observed value of the t-statistic; it's square is equivalent to an F-statistic.
df: The degrees of freedom of the t-statistic.
pval: The p-value for rejecting the NH.
Plots: If argument plots = TRUE, a ggplot object containing empirical operating characteristics corresponding to specified FOM. For example, if FOM = "Wilcoxon" an ROC plot object is produced where reader 1 is CAD. If an LROC FOM is selected, an LROC plot is displayed.

If method = "2T-RRRC" the return value is a list with the following elements:

fomCAD: The observed FOM for CAD.
fomRAD: The observed FOM array for the readers.
avgRadFom: The average FOM of the readers.
avgDiffFom: The mean of the difference FOM, RAD - CAD.
ciDiffFom: A data frame containing the statistics associated with the average difference, RAD - CAD.
ciAvgRdrEachTrt: A data frame containing the statistics associated with the average FOM in each "treatment".
varR: The variance of the pure reader term in the OR model.
varTR: The variance of the treatment-reader term error term in the OR model.
cov1: The covariance1 of the error term - same reader, different treatments.
cov2: The covariance2 of the error term - different readers, same treatment.
cov3: The covariance3 of the error term - different readers, different treatments.
varError: The variance of the pure error term in the OR model.
FStat: The observed value of the F-statistic.
ndf: The numerator degrees of freedom of the F-statistic.
df: The denominator degrees of freedom of the F-statistic.
pval: The p-value for rejecting the NH.
Plots: see above.

Arguments

dataset: The dataset to be analyzed; must be single-treatment at least three readers, where the first reader is CAD.
FOM: The desired FOM; for ROC data it must be "Wilcoxon", for FROC data it can be any valid FOM, e.g., "HrAuc", "wAFROC", etc; for LROC data it must be "Wilcoxon", or "PCL" or "ALROC".
FPFValue: Only needed for LROC data and FOM = "PCL" or "ALROC"; where to evaluate a partial curve based figure of merit. The default is 0.2.
method: The desired analysis: "1T-RRFC","1T-RRRC" (the default) or "2T-RRRC", see manuscript for details.
alpha: Significance level of the test, defaults to 0.05.
plots: Flag, default is FALSE, i.e., a plot is not displayed. If TRUE, it displays the appropriate operating characteristic for all readers and CAD.

Details

PCL is the probability of a correct localization.
The LROC is the plot of PCL (ordinate) vs. FPF.
For LROC data, FOM = "PCL" means the interpolated PCL value at the specified FPFValue.
For FOM = "ALROC" the trapezoidal area under the LROC from FPF = 0 to FPF = FPFValue is used.
If method = "1T-RRRC" the first reader is assumed to be CAD.
If method = "2T-RRRC" the first treatment is assumed to be CAD.
The NH is that the FOM of CAD equals the average of the readers.
The method = "1T-RRRC" analysis uses an adaptation of the single-treatment multiple-reader Obuchowski Rockette (OR) model described in a paper by Hillis (2007), section 5.3. It is characterized by 3 parameters VarR, Var and Cov2, where the latter two are estimated using the jackknife.
For method = "2T-RRRC" the analysis replicates the CAD data as many times as necessary so as to form one "treatment" of an MRMC pairing, the other "treatment" being the radiologists. Then standard ORH analysis is applied. The method is described in Kooi et al. It gives exactly the same final results (F-statistic, ddf and p-value) as "1T-RRRC" but the intermediate quantities are meaningless.

References

Hillis SL (2007) A comparison of denominator degrees of freedom methods for multiple observer ROC studies, Statistics in Medicine. 26:596-619.

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

Hupse R, Samulski M, Lobbes M, et al (2013) Standalone computer-aided detection compared to radiologists performance for the detection of mammographic masses, Eur Radiol. 23(1):93-100.

Kooi T, Gubern-Merida A, et al. (2016) A comparison between a deep convolutional neural network and radiologists for classifying regions of interest in mammography. Paper presented at: International Workshop on Digital Mammography, Malmo, Sweden.

Examples

Run this code

ret1M <- StSignificanceTestingCadVsRad (dataset09, 
FOM = "Wilcoxon", method = "1T-RRRC")

StSignificanceTestingCadVsRad(datasetCadLroc, 
FOM = "Wilcoxon", method = "1T-RRFC")

retLroc1M <- StSignificanceTestingCadVsRad (datasetCadLroc, 
FOM = "PCL", method = "1T-RRRC", FPFValue = 0.05)

## test with fewer readers
dataset09a <- DfExtractDataset(dataset09, rdrs = seq(1:7))
ret1M7 <- StSignificanceTestingCadVsRad (dataset09a, 
FOM = "Wilcoxon", method = "1T-RRRC")

datasetCadLroc7 <- DfExtractDataset(datasetCadLroc, rdrs = seq(1:7))
ret1MLroc7 <- StSignificanceTestingCadVsRad (datasetCadLroc7, 
FOM = "PCL", method = "1T-RRRC", FPFValue = 0.05)

# \donttest{
## takes longer than 5 sec on OSX
## retLroc2M <- StSignificanceTestingCadVsRad (datasetCadLroc, 
## FOM = "PCL", method = "2T-RRRC", FPFValue = 0.05)

## ret2MLroc7 <- StSignificanceTestingCadVsRad (datasetCadLroc7, 
## FOM = "PCL", method = "2T-RRRC", FPFValue = 0.05)
# }

Run the code above in your browser using DataLab