Merge pull request #4 from DIDSR/master

Update my fork

.gitignore

@@ -6,3 +6,10 @@
 /Rpackage/iMRMC.Rcheck
 /Rpackage/iMRMC_1.2.1.tar.gz
 /Rpackage/iMRMC/man
+/Rpackage/iMRMC/input.imrmc
+/Rpackage/iMRMC/tests/testthat/Rplots.pdf
+/iMRMC.Rproj
+/Rpackage/iMRMC_1.2.2.tar.gz
+/Rpackage/iMRMC_1.2.2.zip
+/Rpackage/iMRMC_1.2.3.tar.gz
+/Rpackage/iMRMC_1.2.3.zip

README.md

@@ -1,6 +1,6 @@
 <h3>README for iMRMC and iRoeMetz</h3>
 
-Visit the <a href="https://github.com/DIDSR/iMRMC/releases" rel="nofollow">release Page</a> to download the programs.
+Visit the <a href="https://github.com/DIDSR/iMRMC/releases" rel="nofollow">release Page</a> to download the software or access the repository [HERE](https://github.com/DIDSR/iMRMC).
 
 The documentation for both packages can be found <a href="http://didsr.github.io/iMRMC/" rel="nofollow">here.</a>
 
@@ -15,6 +15,11 @@ Desc: The primary objective of the iMRMC application is to assist investigators
  * The core iMRMC application is a stand-alone, precompiled, license-free Java applications and the source code. It can be used in GUI mode or on the command line.
  * There is also an R package that utilizes the core Java application. Examples for using the programs can be found in the R help files.
 
+Here are some ways to cite our work:
+* Gallas, Brandon D., Andriy Bandos, Frank Samuelson, and Robert F. Wagner. “A Framework for Random-Effects ROC Analysis: Biases with the Bootstrap and Other Variance Estimators.” Commun Stat A-Theory 38, no. 15 (2009): 2586–2603. https://doi.org/10.1080/03610920802610084.
+* Gallas, Brandon D. “One-Shot Estimate of MRMC Variance: AUC.” Acad Radiol 13, no. 3 (2006): 353–62. https://doi.org/10.1016/j.acra.2005.11.030.
+* Gallas, Brandon D. IMRMC v4.0: Application for Analyzing and Sizing MRMC Reader Studies. Silver Spring, MD, 2017. https://github.com/DIDSR/iMRMC/releases, https://cran.r-project.org/web/packages/iMRMC/index.html.
+
 <h5>iRoeMetz</h5>
 
 Desc: The iRoeMetz application can be used to simulate the reader scores for MRMC experiments via Monte Carlo methods given variance components of the ROC scores. The application also estimates the variance components of AUC and can calculate the variance components directly with numerical integration. The simulated experiments can be saved and used in MRMC variance analysis programs, such as iMRMC. The iRoeMetz application is a stand-alone, precompiled, license-free Java applications and the source code.
@@ -23,4 +28,8 @@ Desc: The iRoeMetz application can be used to simulate the reader scores for MRM
 
 Desc: The R package iMRMC executes the iMRMC program which writes the results to the local files system, it reads the analysis results from the local file system, packs the analysis results into a list object, deletes the data and analysis results from the local file system, and returns the list object.
 
+Here is one way to cite our work:
+* Gallas, Brandon D., and Stephen L. Hillis. “Generalized Roe and Metz ROC Model: Analytic Link between Simulated Decision Scores and Empirical AUC Variances and Covariances.” J Med Img 1, no. 3 (2014): 031006. https://doi.org/doi:10.1117/1.JMI.1.3.031006.
+
 [![Github All Releases](https://img.shields.io/github/downloads/DIDSR/iMRMC/total)]()
+

Rpackage/iMRMC/.Rbuildignore

@@ -6,3 +6,5 @@
 ^imrmcDir$
 ^README\.md$
 ^futureVignettes$
+^README-submission-process\.md$
+^cran-check\.md$

Rpackage/iMRMC/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: iMRMC
 Type: Package
 Title: Multi-Reader, Multi-Case Analysis Methods (ROC, Agreement, and Other Metrics)
-Version: 1.2.1
+Version: 1.2.3
 Author: Brandon Gallas
 Maintainer: Brandon Gallas <[email protected]>
 Description:
@@ -23,7 +23,7 @@ License: CC0
 Encoding: UTF-8
 SystemRequirements:	Java JDK 1.7 or higher
 LazyData: true
-Depends: R (>= 2.10)
+Depends: R (>= 3.5.0)
 RoxygenNote: 7.0.2
 Imports: stats
 Suggests: testthat

Rpackage/iMRMC/NEWS.md

@@ -1 +1,3 @@
-Please refer to https://github.com/DIDSR/iMRMC/blob/master/UPDATES_Rpackage_iMRMC.md
+# iMRMC
+
+Information on updates are managed here: https://github.com/DIDSR/iMRMC/blob/master/UPDATES_Rpackage_iMRMC.md

Rpackage/iMRMC/R/iMRMC.R

@@ -116,31 +116,55 @@ doIMRMC <- function(
   iMRMCjarFullPath = NULL,
   stripDatesForTests = FALSE){
 
-  if (is.null(workDir)) {
-    workDir <- normalizePath(tempdir())
+  # If workDir is not specified, it will be the R temp directory.
+  # If workDir is specified, create it.
+  if (is.null(workDir))
+    workDir <- normalizePath(tempdir(), winslash = "/")
+  else
+    dir.create(workDir, showWarnings = FALSE)
+
+  # The inputFile will hold the data for the java program
+  inputFile <- file.path(workDir, "input.imrmc")
+
+  # Check that data is provided
+  if (is.null(data) & is.null(fileName)) {
+    stop("You have not provided an input data frame or an input file.")
+  }
+  # Check that data is not provided twice.
+  if (!is.null(data) & !is.null(fileName)) {
+    stop("You cannot provide an input data frame AND an input file.")
+  }
+
+  # If fileName is specified make sure it exists.
+  # Copy the fileName to the inputFile in the workDir.
+  if (!is.null(fileName)) {
+    if (!file.exists(fileName)) {
+      print(paste("fileName = ", fileName))
+      stop("The full-path fileName provided does not exist.")
+    } 
+    if (fileName == inputFile) {
+      print(paste("fileName = ", fileName))
+      stop("fileName equals the name of the temp inputFile. This is not allowed.")
+    } 
+    file.copy(fileName, inputFile)
   }
 
-  if (is.null(data)) {
-    if (is.null(fileName)) {
-      stop("ERROR: You have not provided an input data frame or an input file.")
-    } else {
-      flagWriteFile <- FALSE
-    }
-  } else {
-    if (!is.null(fileName)) {
-      stop("ERROR: You cannot provide and input data frame AND an input file.")
-    }
-
-    flagWriteFile <- TRUE
-    fileName <- file.path(workDir, "input.imrmc")
+  # Check the input data and write it to the inputFile in the workDir  
+  if (!is.null(data)) {
 
-    # Write data frame to iMRMC input file
-    writeLines("BEGIN DATA:", con = fileName)
-    utils::write.table(data, fileName,
+    # Check that the input data has key columns: readerID, caseID, modalityID, score
+    if (length(setdiff(c("readerID", "caseID", "modalityID", "score"), names(data)))) {
+      stop("The data frame does not include the key columns: readerID, caseID, modalityID, score.")
+    }
+
+    # Write data frame to the inputFile
+    writeLines("BEGIN DATA:", con = inputFile)
+    utils::write.table(data[ , c("readerID", "caseID", "modalityID", "score")], inputFile,
                 quote = FALSE, row.names = FALSE, col.names = FALSE,
                 append = TRUE, sep = ", ")
 
   }
+
   if (is.null(iMRMCjarFullPath)) {
     iMRMCjar <- "iMRMC-v4.0.3.jar"
     pkgPath <- path.package("iMRMC", quiet = FALSE)
@@ -155,30 +179,39 @@ doIMRMC <- function(
 
   # Run iMRMC
   desc <- tryCatch(
-    
+
     system2(
       "java",
-      args = c("-jar",
-               iMRMCjarFullPath,
-               fileName,
-               file.path(workDir, "imrmcDir")),
-      stdout = FALSE, stderr = TRUE),
-
+      args = c(
+        "-jar",
+        iMRMCjarFullPath,
+        inputFile,
+        file.path(workDir, "imrmcDir")
+      ),
+      stdout = TRUE, stderr = TRUE
+    ),
+
     warning = function(w) {
       cat("do_IMRMC WARNING\n")
       warning(w)
     },
+
     error = function(e) {
       cat("\ndoIMRMC ERROR\n")
       cat("One possible reason is that you don't have java.\n")
       cat("This software requires Java JDK 1.7 or higher.\n")
       stop(e)
     }
-
+    
   )
 
+  if (!file.exists(file.path(workDir, "imrmcDir", "AUCperReader.csv"))) {
+    cat(desc, sep = "\n")
+    stop()
+  }
+
   # Retrieve the iMRMC results
-  perReader <- utils::read.csv(file.path(workDir, "imrmcDir", "AUCperReader.csv"))
+  perReader <- utils::read.csv(file.path(workDir, "imrmcDir", "AUCperReader.csv")) 
   Ustat <- utils::read.csv(file.path(workDir, "imrmcDir", "statAnalysis.csv"))
   MLEstat <- utils::read.csv(file.path(workDir, "imrmcDir", "statAnalysisMLE.csv"))
   ROCraw <- utils::read.csv(file.path(workDir, "imrmcDir", "ROCcurves.csv"),
@@ -211,13 +244,9 @@ doIMRMC <- function(
   }
 
   # Delete the content written to disk
-  if (workDir == normalizePath(tempdir())){
+  if (workDir == normalizePath(tempdir(), winslash = "/")) {
     unlink(file.path(workDir, "imrmcDir"), recursive = TRUE)
-
-    if (flagWriteFile) {
-      unlink(fileName)
-    }
-
+    unlink(inputFile)
   }
 
   varDecomp <- list(

Rpackage/iMRMC/R/limitsOfAgreement.R

@@ -62,7 +62,7 @@ laWRBM <- function(
   result2 <- data.frame(
     meanDiff = meanDiff, var.MeanDiff = var.MeanDiff, var.1obs = var.1obs,
     ci95meanDiff.bot = ci95meanDiff.bot, ci95meanDiff.top = ci95meanDiff.top,
-    la.bot = la.bot, la.top = la.top )
+    la.bot = la.bot, la.top = la.top, stringsAsFactors = TRUE)
 
   return(result2)
 
@@ -156,7 +156,7 @@ laBRBM <- function(
   result2 <- data.frame(
     meanDiff = meanDiff, var.MeanDiff = var.MeanDiff, var.1obs = var.Ar1cminusBr2c,
     ci95meanDiff.bot = ci95meanDiff.bot, ci95meanDiff.top = ci95meanDiff.top,
-    la.bot = la.bot, la.top = la.top )
+    la.bot = la.bot, la.top = la.top, stringsAsFactors = TRUE )
 
 return(result2)
 

Rpackage/iMRMC/R/simMRMC.R

@@ -125,7 +125,8 @@ simMRMC <- function(simMRMC.config) {
     readerID = readerID,
     caseID = caseID,
     modalityID = modalityID,
-    score = L_mu + L_cases + L_readers + L_reader.case
+    score = L_mu + L_cases + L_readers + L_reader.case,
+    stringsAsFactors = TRUE
   )
 
 }
@@ -256,7 +257,8 @@ sim.gRoeMetz <- function(config) {
     readerID = rep("-1", nC.neg + nC.pos),
     caseID = c(as.character(caseIDs.neg), as.character(caseIDs.pos)),
     modalityID = rep("truth", nC.neg + nC.pos),
-    score = c(rep(0, nC.neg), rep(1, nC.pos))
+    score = c(rep(0, nC.neg), rep(1, nC.pos)),
+    stringsAsFactors = TRUE
   )
 
   # Simulate the modality independent random effects, negative cases

Rpackage/iMRMC/R/uStats.R

@@ -134,6 +134,13 @@ uStat11.jointD <- function(
   keyColumns = c("readerID", "caseID", "modalityID", "score")
   ) {
 
+  if (class(modalitiesToCompare) != "character") {
+    stop(paste(
+      "class(modalitiesToCompare) =", class(modalitiesToCompare),
+      "... The class should be character.")
+    )
+  }
+
   if (kernelFlag == 1) {
 
     if (length(modalitiesToCompare) != 2) {
@@ -182,7 +189,7 @@ uStat11.jointD <- function(
   sumiD.A <- colSums(design.A)
   sumiD.B <- colSums(design.B)
 
-  meanPerR <- data.frame(sumiS.A / sumiD.A, sumiS.B / sumiD.B)
+  meanPerR <- data.frame(sumiS.A / sumiD.A, sumiS.B / sumiD.B, stringsAsFactors = TRUE)
   names(meanPerR) <- c(desc[1], desc[2])
 
   # Declare the constituent parts of the variance decomposition
@@ -192,7 +199,8 @@ uStat11.jointD <- function(
     c1r0 = c(0, 0, 0),
     c0r1 = c(0, 0, 0),
     c1r1 = c(0, 0, 0),
-    row.names = c("AB", "CD", "ABminusCD")
+    row.names = c("AB", "CD", "ABminusCD"),
+    stringsAsFactors = TRUE
   )
   denom.biased <- numer.biased
 
@@ -291,9 +299,9 @@ uStat11.jointD <- function(
   var.AminusB <- var.A + var.B - 2 * coeff[3, ] %*% moments[3, ]
 
   # Turn these matrices into data frames
-  moments <- data.frame(moments)
+  moments <- data.frame(moments, stringsAsFactors = TRUE)
   names(moments) <- names(numer.biased)
-  coeff <- data.frame(coeff)
+  coeff <- data.frame(coeff, stringsAsFactors = TRUE)
   names(coeff) <- names(numer.biased)
 
   mean <- c(mean.A, mean.B, mean.A - mean.B)
@@ -303,7 +311,7 @@ uStat11.jointD <- function(
   var.1obs <- moments$c1r1 - moments$c0r0
   var.1obs[3] <- var.1obs[1] + var.1obs[2] - 2 * var.1obs[3]
   names(var.1obs) <- desc
-  nCperR <- data.frame(sumiD.A, sumiD.B)
+  nCperR <- data.frame(sumiD.A, sumiD.B, stringsAsFactors = TRUE)
   names(nCperR) <- desc[1:2]
 
   #### Pack Results ####
@@ -339,6 +347,13 @@ uStat11.conditionalD <- function(
   keyColumns = c("readerID", "caseID", "modalityID", "score")
 ) {
 
+  if (class(modalitiesToCompare) != "character") {
+    stop(paste(
+      "class(modalitiesToCompare) =", class(modalitiesToCompare),
+      "... The class should be character.")
+    )
+  }
+
   # Initialize kernel and design matrices ####
   if (kernelFlag == 1) {
 
@@ -390,7 +405,7 @@ uStat11.conditionalD <- function(
   sumiS.A <- colSums(kernel.A)
   sumiS.B <- colSums(kernel.B)
 
-  meanPerR <- data.frame(sumiS.A / sumiD.A, sumiS.B / sumiD.B)
+  meanPerR <- data.frame(sumiS.A / sumiD.A, sumiS.B / sumiD.B, stringsAsFactors = TRUE)
   names(meanPerR) <- c(desc[1], desc[2])
 
   # Reader weights for the reader-averged point estimate
@@ -586,14 +601,16 @@ uStat11.conditionalD <- function(
     c0r0 = c(m.c0r0.AA, m.c0r0.BB, m.c0r0.AB),
     c0r1 = c(m.c0r1.AA, m.c0r1.BB, m.c0r1.AB),
     c1r0 = c(m.c1r0.AA, m.c1r0.BB, m.c1r0.AB),
-    c1r1 = c(m.c1r1.AA, m.c1r1.BB, m.c1r1.AB)
+    c1r1 = c(m.c1r1.AA, m.c1r1.BB, m.c1r1.AB),
+    stringsAsFactors = TRUE
   )
 
   coeff <- data.frame(
     c0r0 = c(c.c0r0.AA, c.c0r0.BB, c.c0r0.AB),
     c0r1 = c(c.c0r1.AA, c.c0r1.BB, c.c0r1.AB),
     c1r0 = c(c.c1r0.AA, c.c1r0.BB, c.c1r0.AB),
-    c1r1 = c(c.c1r1.AA, c.c1r1.BB, c.c1r1.AB)
+    c1r1 = c(c.c1r1.AA, c.c1r1.BB, c.c1r1.AB),
+    stringsAsFactors = TRUE
   )
 
   mean <- c(mean.A, mean.B, mean.A - mean.B)
@@ -603,7 +620,7 @@ uStat11.conditionalD <- function(
   var.1obs <- moments$c1r1 - moments$c0r0
   var.1obs[3] <- var.1obs[1] + var.1obs[2] - 2 * var.1obs[3]
   names(var.1obs) <- desc
-  nCperR <- data.frame(sumiD.A, sumiD.B)
+  nCperR <- data.frame(sumiD.A, sumiD.B, stringsAsFactors = TRUE)
   names(nCperR) <- desc[1:2]
 
   result <- list(
@@ -650,7 +667,8 @@ uStat11.identity <- function(
   df <- data.frame(readerID = factor(df.input[[keyColumns[1]]]),
                    caseID = factor(df.input[[keyColumns[[2]]]]),
                    modalityID = factor(df.input[[keyColumns[[3]]]]),
-                   score = df.input[[keyColumns[[4]]]]
+                   score = df.input[[keyColumns[[4]]]],
+                   stringsAsFactors = TRUE
   )
 
   # Parse out data frames for each modality
@@ -708,7 +726,8 @@ uStat11.diff <- function(
   df <- data.frame(readerID = factor(df.input[[keyColumns[1]]]),
                    caseID = factor(df.input[[keyColumns[[2]]]]),
                    modalityID = factor(df.input[[keyColumns[[3]]]]),
-                   score = df.input[[keyColumns[[4]]]]
+                   score = df.input[[keyColumns[[4]]]],
+                   stringsAsFactors = TRUE
   )
 
   # Parse out data frames for each modality