RJafroc

https://dpc10ster.github.io/RJafroc/
Admin
RJafroc is an R package for analyzing the performance of medical imaging observer studies and...
Tokens:869,846
Snippets:4,550
Trust Score:9.7
Update:5 days ago
Show doc for...
Tokens:
Raw
### Example: SimulateCorCbmDataset()

Source: https://dpc10ster.github.io/RJafroc/reference/SimulateCorCbmDataset.html

Example usage of SimulateCorCbmDataset with default parameters. The commented-out code shows a long-running example with K1=5000 and K2=5000, which is not run by default.

```r
dataset <- SimulateCorCbmDataset()

# \donttest{
## this takes very long
## dataset <- SimulateCorCbmDataset(K1 = 5000, K2 = 5000)
# }
```

--------------------------------

### Example usage of DfCreateCorCbmDataset

Source: https://dpc10ster.github.io/RJafroc/reference/DfCreateCorCbmDataset.html

Creates a dataset with default parameters. The commented lines show how to set a seed and a long-running example.

```R
## seed <- 1 
## this gives unequal numbers of bins in X and Y conditions for 50/50 dataset
dataset <- DfCreateCorCbmDataset()

# \donttest{
## this takes very long time!! used to show asymptotic convergence of ML estimates 
## dataset <- DfCreateCorCbmDataset(K1 = 5000, K2 = 5000)
# }

```

--------------------------------

### Examples: RRRC, FRRC, RRFC power calculations

Source: https://dpc10ster.github.io/RJafroc/reference/SsPowerGivenJK.html

Three examples showing how to compute power for different analysis options (RRRC, FRRC, RRFC) using a pilot dataset and forcing the DBM algorithm with UseDBMHB2004 = TRUE. Each example sets the same parameters but changes analysisOption.

```r
## EXAMPLE 1: RRRC power 
## specify 2-modality ROC dataset and force DBM alg.
res <- SsPowerGivenJK(dataset = dataset02, FOM = "Wilcoxon", effectSize = 0.05, 
J = 6, K = 251, method = "DBM", UseDBMHB2004 = TRUE) # RRRC is default  

## EXAMPLE 1A: FRRC power 
res <- SsPowerGivenJK(dataset = dataset02, FOM = "Wilcoxon", effectSize = 0.05, 
J = 6, K = 251, method = "DBM", UseDBMHB2004 = TRUE, analysisOption = "FRRC") 

## EXAMPLE 1B: RRFC power 
res <- SsPowerGivenJK(dataset = dataset02, FOM = "Wilcoxon", effectSize = 0.05, 
J = 6, K = 251, method = "DBM", UseDBMHB2004 = TRUE, analysisOption = "RRFC")
```

--------------------------------

### Get lesion distribution vector with UtilLesionDistrVector()

Source: https://dpc10ster.github.io/RJafrocQuickStart/quick-start-froc.html

Calls UtilLesionDistrVector(x) to obtain the lesion distribution vector. The function returns an array with two columns: the first column contains the distinct non-zero lesion counts (14 in this example), and the second column contains the fraction of diseased cases with that number of lesions. The second column sums to unity.

```R
lesDistr <- UtilLesionDistrVector(x)
lesDistr
#>  [1] 0.217391304 0.200000000 0.113043478 0.086956522 0.043478261 0.095652174
#>  [7] 0.052173913 0.069565217 0.017391304 0.026086957 0.026086957 0.026086957
#> [13] 0.017391304 0.008695652
```

--------------------------------

### Example: OR-based algorithm with pilot dataset

Source: https://dpc10ster.github.io/RJafroc/reference/SsPowerGivenJK.html

A simple example using the default OR method with a pilot dataset. This is the default behavior when method is not specified.

```r
## EXAMPLE 3: specify 2-modality ROC dataset and use OR-based alg.
res <- SsPowerGivenJK(dataset = dataset02, FOM = "Wilcoxon", effectSize = 0.05, 
J = 6, K = 251)
```

--------------------------------

### St usage examples

Source: https://dpc10ster.github.io/RJafroc/reference/St.html

Demonstrates calling St with different datasets, FOMs, and methods. The first three lines show basic usage; the last two are wrapped in \donttest{} to avoid running in examples.

```r
result <- St(dataset02,FOM = "Wilcoxon", method = "DBM")
result <- St(dataset02,FOM = "Wilcoxon", method = "OR")
result <- St(datasetX, FOM = "wAFROC", method = "OR", analysisOption = "RRRC")

# \donttest{
result <- St(dataset05, FOM = "wAFROC")
result <- St(dataset05, FOM = "HrAuc", method = "DBM")
# }
```

--------------------------------

### Example: Save dataset05 to rocData2.xlsx

Source: https://dpc10ster.github.io/RJafroc/reference/DfWriteExcelDataFile.html

Example call to DfWriteExcelDataFile that saves the dataset05 object to an Excel file named 'rocData2.xlsx'. The example is wrapped in \donttest{} and commented out with #.

```r
# \donttest{
##DfWriteExcelDataFile(dataset = dataset05, fileName = "rocData2.xlsx")
# }
```

--------------------------------

### Test RJafroc installation with PlotCbmFit

Source: https://dpc10ster.github.io/RJafroc/index.html

Run this R code after installing RJafroc to verify the installation. It creates a CBM fit plot and prints it, which should display two ROC plots in the Plots window.

```R
cbmPlot <- PlotCbmFit(c(1, 2), c(0.5, 0.5))
print(cbmPlot)
```

--------------------------------

### SimulateFrocDataset example with plotting

Source: https://dpc10ster.github.io/RJafroc/reference/SimulateFrocDataset.html

Example that sets parameters, calls SimulateFrocDataset to generate a dataset, and then plots empirical FROC operating characteristics using PlotEmpOpChrs. The seed is set to 1 for reproducibility.

```r
set.seed(1) 
K1 <- 5;K2 <- 7;
maxLL <- 2;perCase <- floor(runif(K2, 1, maxLL + 1))
mu <- 1;lambda <- 1;nu <- 0.99 ;zeta1 <- -1
I <- 2; J <- 5

frocDataRaw <- SimulateFrocDataset(
  mu = mu, lambda = lambda, nu = nu, zeta1 = zeta1,
  I = I, J = J, K1 = K1, K2 = K2, perCase = perCase )
  
## plot the data
ret <- PlotEmpOpChrs(frocDataRaw, opChType = "FROC")
## print(ret$Plot)
```

--------------------------------

### RSM_wLLF example call

Source: https://dpc10ster.github.io/RJafroc/reference/RSM_wLLF.html

Example call to RSM_wLLF with sample parameters. The output is 0.34174.

```r
RSM_wLLF(1, 1, 0.9, lesDistr = c(0.5, 0.4, 0.1), relWeights = c(0.7, 0.2, 0.1)) 
#> [1] 0.34174
## 0.34174
```

--------------------------------

### R example: run doBootstrap with parametricFOM=TRUE

Source: https://dpc10ster.github.io/RJafrocBook/sources-of-variability.html

Sets up the input data and calls `doBootstrap` with `parametricFOM=TRUE`, `B=200`, `seed=1`. The output shows the original AUC, mean bootstrap AUC, and standard deviation. The example uses a 2x5 counts table.

```R
parametricFOM <- TRUE
B <- 200;seed <- 1;set.seed(seed)
RocTable = array(dim = c(2,5))
RocTable[1,]  <- c(30,19,8,2,1)
RocTable[2,]  <- c(5,6,5,12,22)

ret <- doBootstrap(parametricFOM, B, seed, RocTable)
OrigAUC <- ret$OrigAUC
meanAUCboot <- ret$meanAUCboot
stdAUCboot <- ret$stdAUCboot

cat("Bootstrap variance estimation:",
    "\nparametricFOM = ", parametricFOM,
    "\nseed = ", seed, 
    "\nB = ", B,
    "\nOrigAUC = ", OrigAUC, 
    "\nmeanAUCboot = ", meanAUCboot, 
    "\nstdAUCboot = ", stdAUCboot, "\n")
```

```R
#> Bootstrap variance estimation: 
#> parametricFOM =  TRUE 
#> seed =  1 
#> B =  200 
#> OrigAUC =  0.8704519 
#> meanAUCboot =  0.8671713 
#> stdAUCboot =  0.04380523
```

--------------------------------

### SsFrocSampleSize example with dataset04

Source: https://dpc10ster.github.io/RJafroc/reference/SsFrocSampleSize.html

Example call to SsFrocSampleSize using a subset of dataset04, with effectSizeROC 0.03, 5 readers, 100 cases, and a lesion distribution. The example is wrapped in \donttest{} and includes timing output.

```r
# \donttest{
## Examples with CPU or elapsed time > 5s
## user system elapsed
## SsFrocSampleSize 8.102  0.023   8.135

## SsFrocSampleSize(DfExtractDataset(dataset04, trts = c(1,2)), 
## effectSizeROC = 0.03, JPivot = 5, KPivot = 100, lesDistr = c(0.69, 0.2, 0.11))
# }
```

--------------------------------

### Example: NULL dataset with DBM variance components

Source: https://dpc10ster.github.io/RJafroc/reference/SsPowerGivenJK.html

Shows how to supply variance components when dataset is NULL. The example uses UtilDBMVarComp to obtain variance components and passes them as a list to the ... argument. Note the comments indicate that the right-hand side values should be replaced with actual values.

```r
## EXAMPLE 2: specify NULL dataset & DBM var. comp. & force DBM-based alg.
vcDBM <- UtilDBMVarComp(dataset02, FOM = "Wilcoxon")$VarCom
res <- SsPowerGivenJK(dataset = NULL, FOM = "Wilcoxon", J = 6, K = 251, 
effectSize = 0.05, method = "DBM", UseDBMHB2004 = TRUE, 
list( 
VarTR = vcDBM["VarTR","Estimates"], # replace rhs with actual values as in 4A
VarTC = vcDBM["VarTC","Estimates"], # do:
VarErr = vcDBM["VarErr","Estimates"])) # do:
```
Tokens:
Raw