diff --git a/NAMESPACE b/NAMESPACE
index 8cab948..7f72bd5 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -136,6 +136,7 @@ export(getMixtureDensityLossFunction)
 export(getMixtureDensityMseAccuracyFunction)
 export(getMixtureDensitySamplingFunction)
 export(getPretrainedNetwork)
+export(harvardOxfordAtlasLabeling)
 export(hippMapp3rSegmentation)
 export(histogramWarpImageIntensities)
 export(hyperMapp3rSegmentation)
diff --git a/R/brainExtraction.R b/R/brainExtraction.R
index dc293bd..d529af7 100644
--- a/R/brainExtraction.R
+++ b/R/brainExtraction.R
@@ -15,7 +15,7 @@
 #'   \item{"t1combined": }{Brian's combination of "t1" and "t1nobrainer".  One can also specify
 #'                         "t1combined[X]" where X is the morphological radius.  X = 12 by default.}
 #'   \item{"t1threetissue": }{T1-weighted MRI---originally developed from BrainWeb20 (and later expanded). 
-#'                        Label 1:  parenchyma, label 2: meninges/csf, label 3: misc. head.
+#'                        Label 1:  parenchyma, label 2: meninges/csf, label 3: misc. head.}
 #'   \item{"flair": }{FLAIR MRI.}
 #'   \item{"t2": }{T2-w MRI.}
 #'   \item{"bold": }{3-D mean BOLD MRI.}
diff --git a/R/desikanKillianyTourvilleLabeling.R b/R/desikanKillianyTourvilleLabeling.R
index 541c2fb..217058c 100644
--- a/R/desikanKillianyTourvilleLabeling.R
+++ b/R/desikanKillianyTourvilleLabeling.R
@@ -234,7 +234,7 @@ desikanKillianyTourvilleLabeling <- function( t1, doPreprocessing = TRUE,
   #
   ################################
 
-  t1Preprocessed <- t1
+  t1Preprocessed <- antsImageClone( t1 )
   if( doPreprocessing )
     {
     t1Preprocessing <- preprocessBrainImage( t1,
diff --git a/R/getPretrainedNetwork.R b/R/getPretrainedNetwork.R
index 359925f..b76dd98 100644
--- a/R/getPretrainedNetwork.R
+++ b/R/getPretrainedNetwork.R
@@ -82,6 +82,7 @@ getPretrainedNetwork <- function(
               "dktInner",
               "dktOuter",
               "dktOuterWithSpatialPriors",
+              "HarvardOxfordAtlasSubcortical",
               "ex5_coronal_weights",
               "ex5_sagittal_weights",
               "allen_brain_mask_weights",
@@ -154,7 +155,7 @@ getPretrainedNetwork <- function(
               "wholeHeadInpaintingFLAIR",
               "wholeHeadInpaintingPatchBasedT1",
               "wholeHeadInpaintingPatchBasedFLAIR",
-	             "wholeTumorSegmentationT2Flair",
+	            "wholeTumorSegmentationT2Flair",
               "wholeLungMaskFromVentilation",
               "DeepAtroposHcpT1Weights",
               "DeepAtroposHcpT1T2Weights",
@@ -237,6 +238,7 @@ getPretrainedNetwork <- function(
     dktInner = "https://ndownloader.figshare.com/files/23266943",
     dktOuter = "https://ndownloader.figshare.com/files/23765132",
     dktOuterWithSpatialPriors = "https://ndownloader.figshare.com/files/24230768",
+    HarvardOxfordAtlasSubcortical = "https://figshare.com/ndownloader/files/50577546",
     e13x5_coronal_weights = "https://figshare.com/ndownloader/files/38035968",
     e13x5_sagittal_weights = "https://figshare.com/ndownloader/files/38035965",
     allen_brain_mask_weights = "https://figshare.com/ndownloader/files/36999880",
diff --git a/R/harvardOxfordAtlasLabeling.R b/R/harvardOxfordAtlasLabeling.R
new file mode 100644
index 0000000..bd72951
--- /dev/null
+++ b/R/harvardOxfordAtlasLabeling.R
@@ -0,0 +1,223 @@
+#' Subcortical and cerebellar labeling from a T1 image.
+#'
+#' Perform HOA labeling using deep learning and data from 
+# "High Resolution, Comprehensive Atlases of the Human Brain 
+#' Morphology" number: "NIH NIMH 5R01MH112748-04". Repository: 
+#' https://github.com/HOA-2/SubcorticalParcellations'
+#'
+#' The labeling is as follows:
+#' \itemize{
+#' \item{Label 1:}{Lateral Ventricle Left}
+#' \item{Label 2:}{Lateral Ventricle Right}
+#' \item{Label 3:}{CSF}
+#' \item{Label 4:}{Third Ventricle}
+#' \item{Label 5:}{Fourth Ventricle}
+#' \item{Label 6:}{5th Ventricle}
+#' \item{Label 7:}{Nucleus Accumbens Left}
+#' \item{Label 8:}{Nucleus Accumbens Right}
+#' \item{Label 9:}{Caudate Left}
+#' \item{Label 10:}{Caudate Right}
+#' \item{Label 11:}{Putamen Left}
+#' \item{Label 12:}{Putamen Right}
+#' \item{Label 13:}{Globus Pallidus Left}
+#' \item{Label 14:}{Globus Pallidus Right}
+#' \item{Label 15:}{Brainstem}
+#' \item{Label 16:}{Thalamus Left}
+#' \item{Label 17:}{Thalamus Right}
+#' \item{Label 18:}{Inferior Horn of the Lateral Ventricle Left}
+#' \item{Label 19:}{Inferior Horn of the Lateral Ventricle Right}
+#' \item{Label 20:}{Hippocampal Formation Left}
+#' \item{Label 21:}{Hippocampal Formation Right}
+#' \item{Label 22:}{Amygdala Left}
+#' \item{Label 23:}{Amygdala Right}
+#' \item{Label 24:}{Optic Chiasm}
+#' \item{Label 25:}{VDC Anterior Left}
+#' \item{Label 26:}{VDC Anterior Right}
+#' \item{Label 27:}{VDC Posterior Left}
+#' \item{Label 28:}{VDC Posterior Right}
+#' \item{Label 29:}{Cerebellar Cortex Left}
+#' \item{Label 30:}{Cerebellar Cortex Right}
+#' \item{Label 31:}{Cerebellar White Matter Left}
+#' \item{Label 32:}{Cerebellar White Matter Right}
+#' }
+#'
+#' Preprocessing on the training data consisted of:
+#'    * n4 bias correction,
+#'    * brain extraction, and
+#'    * affine registration to HCP.
+#' The input T1 should undergo the same steps.  If the input T1 is the raw
+#' T1, these steps can be performed by the internal preprocessing, i.e. set
+#' \code{doPreprocessing = TRUE}
+#'
+#' @param t1 raw or preprocessed 3-D T1-weighted brain image.
+#' @param doPreprocessing perform preprocessing.  See description above.
+#' @param verbose print progress.
+#' @return list consisting of the segmentation image and probability images for
+#' each label.
+#' @author Tustison NJ
+#' @examples
+#' \dontrun{
+#' library( ANTsRNet )
+#' library( keras )
+#'
+#' image <- antsImageRead( "t1.nii.gz" )
+#' results <- harvardOxfordAtlasLabeling( image )
+#' }
+#' @export
+harvardOxfordAtlasLabeling <- function( t1, doPreprocessing = TRUE,
+  verbose = FALSE )
+{
+
+  if( t1@dimension != 3 )
+    {
+    stop( "Image dimension must be 3." )
+    }
+
+  reshapeImage <- function( image, cropSize, interpType = "linear" )
+    {
+    imageResampled <- NULL
+    if( interpType == "linear" )
+      {
+      imageResampled <- resampleImage( image, c( 1, 1, 1 ), useVoxels = FALSE, interpType = 0 )
+      } else {
+      imageResampled <- resampleImage( image, c( 1, 1, 1 ), useVoxels = FALSE, interpType = 1 )
+      }
+    imageCropped <- padOrCropImageToSize( imageResampled, cropSize ) 
+    return( imageCropped )
+    }
+
+  whichTemplate <- "hcpyaT1Template"
+  templateTransformType <- "antsRegistrationSyNQuick[a]"
+  template <- antsImageRead( getANTsXNetData( whichTemplate ) )
+
+  croppedTemplateSize <- c( 160, 176, 160 )
+
+  ################################
+  #
+  # Preprocess image
+  #
+  ################################
+
+  t1Preprocessed <- antsImageClone( t1 )
+  if( doPreprocessing )
+    {
+    t1Preprocessing <- preprocessBrainImage( t1,
+        truncateIntensity = NULL,
+        brainExtractionModality = "t1threetissue",
+        template = whichTemplate,
+        templateTransformType = templateTransformType,
+        doBiasCorrection = TRUE,
+        doDenoising = FALSE,
+        verbose = verbose )
+    t1Preprocessed <- t1Preprocessing$preprocessedImage * t1Preprocessing$brainMask
+    t1Preprocessed <- reshapeImage( t1Preprocessed, cropSize = croppedTemplateSize )
+    }
+
+  ################################
+  #
+  # Build model and load weights
+  #
+  ################################
+
+  labels <- 0:35
+  channelSize <- 1
+  numberOfClassificationLabels <- length( labels )
+
+  unetModelPre <- createUnetModel3D( c( croppedTemplateSize, channelSize ),
+    numberOfOutputs = numberOfClassificationLabels, mode = 'classification',
+    numberOfFilters = c( 16, 32, 64, 128 ), dropoutRate = 0.0,
+    convolutionKernelSize = c( 3, 3, 3 ), deconvolutionKernelSize = c( 2, 2, 2 ),
+    weightDecay = 0.0 )
+
+  penultimateLayer <- unetModelPre$layers[[length( unetModelPre$layers ) - 1]]$output
+
+  output2 <-  penultimateLayer %>%
+             keras::layer_conv_3d( filters = 1, 
+                                   kernel_size = c( 1, 1, 1 ),
+                                   activation = 'sigmoid',
+                                   kernel_regularizer = keras::regularizer_l2( 0.0 ) )
+
+  unetModel <- keras::keras_model( inputs = unetModelPre$input, 
+                                   outputs = list( unetModelPre$output, output2 ) )
+  weightsFileNamePath <- getPretrainedNetwork( "HarvardOxfordAtlasSubcortical" )
+  keras::load_model_weights_hdf5( unetModel, filepath = weightsFileNamePath )
+
+  ################################
+  #
+  # Do prediction and normalize to native space
+  #
+  ################################
+
+  if( verbose )
+    {
+    cat( "Model prediction using both the original and contralaterally flipped version\n" )
+    }
+
+  batchX <- array( data = 0, dim = c( 2, croppedTemplateSize, channelSize ) )
+  batchX[1,,,,1] <- as.array( iMath( t1Preprocessed, "Normalize" ) )
+  batchX[2,,,,1] <- batchX[1,croppedTemplateSize[1]:1,,,1] 
+
+  predictedData <- unetModel %>% predict( batchX, verbose = verbose )
+
+  probabilityImages <- list()
+
+  hoaLateralLabels <- c( 0, 3, 4, 5, 6, 15, 24 )
+  hoaLateralLeftLabels <- c( 1, 7, 9, 11, 13, 16, 18, 20, 22, 25, 27, 29, 31 )
+  hoaLateralRightLabels <- c( 2, 8, 10, 12, 14, 17, 19, 21, 23, 26, 28, 30, 32 )
+
+  hoaLabels <- list()
+  hoaLabels[[1]] <- hoaLateralLabels
+  hoaLabels[[2]] <- hoaLateralLeftLabels
+  hoaLabels[[3]] <- hoaLateralRightLabels
+
+  for( b in seq.int( 2 ) )
+    {
+    for( i in seq.int( length( hoaLabels ) ) ) 
+      {
+      for( j in seq.int( length( hoaLabels[[i]] ) ) )
+        {
+        label <- hoaLabels[[i]][j]   
+        probabilityArray <- drop( predictedData[[1]][b,,,,label+1] )
+        if( label == 0 )
+          {
+          probabilityArray <- probabilityArray + drop( rowSums( predictedData[[1]][b,,,,34:36, drop=FALSE] , dims = 4 ) )
+          }
+        if( b == 2 )
+          {
+          probabilityArray <- probabilityArray[dim( probabilityArray )[1]:1,,]
+          if( i == 2 )
+            {
+            label <- hoaLateralRightLabels[j]
+            } else if( i == 3 ) {
+            label <- hoaLateralLeftLabels[j]
+            }
+          }
+        probabilityImage <- as.antsImage( probabilityArray, reference = t1Preprocessed )
+        if( doPreprocessing )
+          {
+          probabilityImage <- padOrCropImageToSize( probabilityImage, dim( template ) )
+          probabilityImage <- antsApplyTransforms( fixed = t1, moving = probabilityImage,
+                  transformlist = t1Preprocessing$templateTransforms$invtransforms,
+                    whichtoinvert = c( TRUE ), interpolator = "linear", verbose = verbose )
+          }
+        if( b == 1 )  
+          {
+          probabilityImages[[label + 1]] <- probabilityImage
+          } else {
+          probabilityImages[[label + 1]] <- 0.5 * ( probabilityImages[[label + 1]] + probabilityImage )
+          }
+        }
+      }
+    }
+
+  imageMatrix <- imageListToMatrix( probabilityImages, t1 * 0 + 1 )
+  segmentationMatrix <- matrix( apply( imageMatrix, 2, which.max ), nrow = 1 )
+  segmentationImage <- matrixToImages( segmentationMatrix, t1 * 0 + 1 )[[1]] - 1
+
+  return( list(
+          segmentationImage = segmentationImage,
+          probabilityImages = probabilityImages
+          )
+        )
+}
+
diff --git a/man/brainExtraction.Rd b/man/brainExtraction.Rd
index 348eb3f..f132cf3 100644
--- a/man/brainExtraction.Rd
+++ b/man/brainExtraction.Rd
@@ -6,9 +6,9 @@
 \usage{
 brainExtraction(
   image,
-  modality = c("t1", "t1.v0", "t1.v1", "t1nobrainer", "t1combined", "t2", "t2.v0",
-    "t2star", "flair", "flair.v0", "bold", "bold.v0", "fa", "fa.v0", "mra", "t1t2infant",
-    "t1infant", "t2infant"),
+  modality = c("t1", "t1.v0", "t1.v1", "t1nobrainer", "t1combined", "t1threetissue",
+    "t2", "t2.v0", "t2star", "flair", "flair.v0", "bold", "bold.v0", "fa", "fa.v0",
+    "mra", "t1t2infant", "t1infant", "t2infant"),
   verbose = FALSE
 )
 }
@@ -21,6 +21,8 @@ brainExtraction(
 \item{"t1nobrainer": }{T1-weighted MRI---FreeSurfer-trained: h/t Satra Ghosh and Jakub Kaczmarzyk.}
 \item{"t1combined": }{Brian's combination of "t1" and "t1nobrainer".  One can also specify
 "t1combined\link{X}" where X is the morphological radius.  X = 12 by default.}
+\item{"t1threetissue": }{T1-weighted MRI---originally developed from BrainWeb20 (and later expanded).
+Label 1:  parenchyma, label 2: meninges/csf, label 3: misc. head.}
 \item{"flair": }{FLAIR MRI.}
 \item{"t2": }{T2-w MRI.}
 \item{"bold": }{3-D mean BOLD MRI.}
diff --git a/man/createDenseNetModel2D.Rd b/man/createDenseNetModel2D.Rd
index 928bb98..b3349ce 100644
--- a/man/createDenseNetModel2D.Rd
+++ b/man/createDenseNetModel2D.Rd
@@ -12,7 +12,7 @@ createDenseNetModel2D(
   numberOfDenseBlocks = 1,
   growthRate = 12,
   dropoutRate = 0.2,
-  weightDecay = 1e-04,
+  weightDecay = 0.0001,
   mode = "classification"
 )
 }
diff --git a/man/createDenseNetModel3D.Rd b/man/createDenseNetModel3D.Rd
index 1722887..5f5c346 100644
--- a/man/createDenseNetModel3D.Rd
+++ b/man/createDenseNetModel3D.Rd
@@ -12,7 +12,7 @@ createDenseNetModel3D(
   numberOfDenseBlocks = 1,
   growthRate = 12,
   dropoutRate = 0.2,
-  weightDecay = 1e-04,
+  weightDecay = 0.0001,
   mode = "classification"
 )
 }
diff --git a/man/createDenseUnetModel2D.Rd b/man/createDenseUnetModel2D.Rd
index d41d5e4..83edbf9 100644
--- a/man/createDenseUnetModel2D.Rd
+++ b/man/createDenseUnetModel2D.Rd
@@ -13,7 +13,7 @@ createDenseUnetModel2D(
   reductionRate = 0,
   depth = 7,
   dropoutRate = 0,
-  weightDecay = 1e-04,
+  weightDecay = 0.0001,
   mode = c("classification", "regression")
 )
 }
diff --git a/man/createDenseUnetModel3D.Rd b/man/createDenseUnetModel3D.Rd
index d422de5..6fa3426 100644
--- a/man/createDenseUnetModel3D.Rd
+++ b/man/createDenseUnetModel3D.Rd
@@ -13,7 +13,7 @@ createDenseUnetModel3D(
   reductionRate = 0,
   depth = 7,
   dropoutRate = 0,
-  weightDecay = 1e-04,
+  weightDecay = 0.0001,
   mode = c("classification", "regression")
 )
 }
diff --git a/man/createResUnetModel2D.Rd b/man/createResUnetModel2D.Rd
index 4282c4f..445b427 100644
--- a/man/createResUnetModel2D.Rd
+++ b/man/createResUnetModel2D.Rd
@@ -12,7 +12,7 @@ createResUnetModel2D(
   convolutionKernelSize = c(3, 3),
   deconvolutionKernelSize = c(2, 2),
   dropoutRate = 0,
-  weightDecay = 1e-04,
+  weightDecay = 0.0001,
   mode = c("classification", "regression")
 )
 }
diff --git a/man/createResUnetModel3D.Rd b/man/createResUnetModel3D.Rd
index 4b7b0f2..15f4721 100644
--- a/man/createResUnetModel3D.Rd
+++ b/man/createResUnetModel3D.Rd
@@ -12,7 +12,7 @@ createResUnetModel3D(
   convolutionKernelSize = c(3, 3, 3),
   deconvolutionKernelSize = c(2, 2, 2),
   dropoutRate = 0,
-  weightDecay = 1e-04,
+  weightDecay = 0.0001,
   mode = c("classification", "regression")
 )
 }
diff --git a/man/createRmnetGenerator.Rd b/man/createRmnetGenerator.Rd
index e2239db..3c681d0 100644
--- a/man/createRmnetGenerator.Rd
+++ b/man/createRmnetGenerator.Rd
@@ -4,7 +4,7 @@
 \alias{createRmnetGenerator}
 \title{Implementation of the "RMNet" generator architecture for inpainting}
 \usage{
-createRmnetGenerator()
+createRmnetGenerator(numberOfChannels = 3)
 }
 \value{
 a keras model
diff --git a/man/createShivaUnetModel3D.Rd b/man/createShivaUnetModel3D.Rd
index b8b2a4d..31033f4 100644
--- a/man/createShivaUnetModel3D.Rd
+++ b/man/createShivaUnetModel3D.Rd
@@ -5,11 +5,14 @@
 \title{Implementation of the "shiva" u-net architecture for PVS and WMH
 segmentation}
 \usage{
-createShivaUnetModel3D(numberOfModalities = 1)
+createShivaUnetModel3D(numberOfModalities = 1, numberOfOutputs = 1)
 }
 \arguments{
 \item{numberOfModalities}{Specifies number of channels in the
 architecture.}
+
+\item{numberOfOutputs}{Specifies the number of outputs per voxel.
+Determines final activation function (1 = sigmoid, >1 = softmax).}
 }
 \value{
 a u-net keras model
diff --git a/man/createSsdModel2D.Rd b/man/createSsdModel2D.Rd
index 570cc7e..399550d 100644
--- a/man/createSsdModel2D.Rd
+++ b/man/createSsdModel2D.Rd
@@ -7,7 +7,7 @@
 createSsdModel2D(
   inputImageSize,
   numberOfOutputs,
-  l2Regularization = 5e-04,
+  l2Regularization = 0.0005,
   minScale = 0.1,
   maxScale = 0.9,
   aspectRatiosPerLayer = list(c("1:1", "2:1", "1:2"), c("1:1", "2:1", "1:2", "3:1",
diff --git a/man/createSsdModel3D.Rd b/man/createSsdModel3D.Rd
index 92de21e..969b93e 100644
--- a/man/createSsdModel3D.Rd
+++ b/man/createSsdModel3D.Rd
@@ -7,7 +7,7 @@
 createSsdModel3D(
   inputImageSize,
   numberOfOutputs,
-  l2Regularization = 5e-04,
+  l2Regularization = 0.0005,
   minScale = 0.1,
   maxScale = 0.9,
   aspectRatiosPerLayer = list(c("1:1:1", "2:1:1", "1:2:1", "1:1:2"), c("1:1:1", "2:1:1",
diff --git a/man/createWideResNetModel2D.Rd b/man/createWideResNetModel2D.Rd
index 1e7509e..2a1d2e1 100644
--- a/man/createWideResNetModel2D.Rd
+++ b/man/createWideResNetModel2D.Rd
@@ -12,7 +12,7 @@ createWideResNetModel2D(
   residualBlockSchedule = c(16, 32, 64),
   poolSize = c(8, 8),
   dropoutRate = 0,
-  weightDecay = 5e-04,
+  weightDecay = 0.0005,
   mode = "classification"
 )
 }
diff --git a/man/createWideResNetModel3D.Rd b/man/createWideResNetModel3D.Rd
index 5f0fa9e..07cf6ac 100644
--- a/man/createWideResNetModel3D.Rd
+++ b/man/createWideResNetModel3D.Rd
@@ -12,7 +12,7 @@ createWideResNetModel3D(
   residualBlockSchedule = c(16, 32, 64),
   poolSize = c(8, 8, 8),
   dropoutRate = 0,
-  weightDecay = 5e-04,
+  weightDecay = 0.0005,
   mode = c("classification", "regression")
 )
 }
diff --git a/man/getANTsXNetData.Rd b/man/getANTsXNetData.Rd
index 007134d..c7cf7f9 100644
--- a/man/getANTsXNetData.Rd
+++ b/man/getANTsXNetData.Rd
@@ -23,7 +23,9 @@ getANTsXNetData(
     "DevCCF_P04_STPT_50um_BrainParcellationJayMask", 
      "hcpyaT1Template",
     "hcpyaT2Template", "hcpyaFATemplate", "hcpyaTemplateBrainMask",
-    "hcpyaTemplateBrainSegmentation"),
+    "hcpyaTemplateBrainSegmentation", "hcpinterT1Template", "hcpinterT2Template",
+    "hcpinterFATemplate", "hcpinterTemplateBrainMask",
+    "hcpinterTemplateBrainSegmentation"),
   targetFileName
 )
 }
diff --git a/man/getPretrainedNetwork.Rd b/man/getPretrainedNetwork.Rd
index b0b3d66..3c96381 100644
--- a/man/getPretrainedNetwork.Rd
+++ b/man/getPretrainedNetwork.Rd
@@ -14,53 +14,54 @@ getPretrainedNetwork(
     "brainExtractionRobustFA", "brainExtractionNoBrainer", 
     
     "brainExtractionInfantT1T2", "brainExtractionInfantT1", "brainExtractionInfantT2",
-    "brainSegmentation", "brainSegmentationPatchBased", "bratsStage1", "bratsStage2",
-    "cerebellumWhole", "cerebellumTissue", "cerebellumLabels", "claustrum_axial_0",
-    "claustrum_axial_1", "claustrum_axial_2", "claustrum_coronal_0",
-    "claustrum_coronal_1", "claustrum_coronal_2", "ctHumanLung", "dbpn4x", "deepFlash",
-    "deepFlashLeftT1", "deepFlashRightT1", "deepFlashLeftBoth", "deepFlashRightBoth",
-    "deepFlashLeftT1Hierarchical", 
-     "deepFlashRightT1Hierarchical",
-    "deepFlashLeftBothHierarchical", "deepFlashRightBothHierarchical",
-    "deepFlashLeftT1Hierarchical_ri", "deepFlashRightT1Hierarchical_ri",
-    "deepFlashLeftBothHierarchical_ri", "deepFlashRightBothHierarchical_ri",
-    "deepFlash2LeftT1Hierarchical", "deepFlash2RightT1Hierarchical", "deepFlashLeft8",
-    "deepFlashRight8", "deepFlashLeft16", "deepFlashRight16", "deepFlashLeft16new",
-    "deepFlashRight16new", "denoising", "dktInner", "dktOuter",
-    "dktOuterWithSpatialPriors", "ex5_coronal_weights", 
-     "ex5_sagittal_weights",
+    "brainExtractionBrainWeb20", "brainSegmentation", "brainSegmentationPatchBased",
+    "bratsStage1", "bratsStage2", "cerebellumWhole", "cerebellumTissue",
+    "cerebellumLabels", "claustrum_axial_0", "claustrum_axial_1", "claustrum_axial_2",
+    "claustrum_coronal_0", "claustrum_coronal_1", "claustrum_coronal_2", "ctHumanLung",
+    "dbpn4x", "deepFlash", "deepFlashLeftT1", "deepFlashRightT1", "deepFlashLeftBoth",
+    "deepFlashRightBoth", 
+     "deepFlashLeftT1Hierarchical",
+    "deepFlashRightT1Hierarchical", "deepFlashLeftBothHierarchical",
+    "deepFlashRightBothHierarchical", "deepFlashLeftT1Hierarchical_ri",
+    "deepFlashRightT1Hierarchical_ri", "deepFlashLeftBothHierarchical_ri",
+    "deepFlashRightBothHierarchical_ri", "deepFlash2LeftT1Hierarchical",
+    "deepFlash2RightT1Hierarchical", "deepFlashLeft8", "deepFlashRight8",
+    "deepFlashLeft16", "deepFlashRight16", "deepFlashLeft16new", "deepFlashRight16new",
+    "denoising", "dktInner", "dktOuter", "dktOuterWithSpatialPriors", 
+    
+    "HarvardOxfordAtlasSubcortical", "ex5_coronal_weights", "ex5_sagittal_weights",
     "allen_brain_mask_weights", "functionalLungMri", "hippMapp3rInitial",
     "hippMapp3rRefine", "hyperMapp3r", "hypothalamus",
     "inpainting_sagittal_rmnet_weights", "inpainting_coronal_rmnet_weights",
     "inpainting_axial_rmnet_weights", "inpainting_axial_rmnet_flair_weights",
     "inpainting_coronal_rmnet_flair_weights", "inpainting_sagittal_rmnet_flair_weights",
-    "koniqMBCS", "koniqMS", "koniqMS2", "koniqMS3", "lesion_whole_brain",
-    "lungCtWithPriorsSegmentationWeights", "maskLobes", 
-     "mriSuperResolution",
+    "koniqMBCS", "koniqMS", "koniqMS2", "koniqMS3", "lesion_whole_brain", 
+    
+    "lungCtWithPriorsSegmentationWeights", "maskLobes", "mriSuperResolution",
     "mriModalityClassification", "mraVesselWeights_160", "mouseMriBrainExtraction",
     "mouseT2wBrainExtraction3D", "mouseT2wBrainParcellation3DNick",
     "mouseT2wBrainParcellation3DTct", "mouseSTPTBrainParcellation3DJay",
     "pvs_shiva_t1_0", "pvs_shiva_t1_1", "pvs_shiva_t1_2", "pvs_shiva_t1_3",
     "pvs_shiva_t1_4", "pvs_shiva_t1_5", "pvs_shiva_t1_flair_0", "pvs_shiva_t1_flair_1",
-    "pvs_shiva_t1_flair_2", "pvs_shiva_t1_flair_3", "pvs_shiva_t1_flair_4",
-    "wmh_shiva_flair_0", "wmh_shiva_flair_1", 
-     "wmh_shiva_flair_2",
-    "wmh_shiva_flair_3", "wmh_shiva_flair_4", "wmh_shiva_t1_flair_0",
-    "wmh_shiva_t1_flair_1", "wmh_shiva_t1_flair_2", "wmh_shiva_t1_flair_3",
-    "wmh_shiva_t1_flair_4", "protonLungMri", "protonLobes", "pulmonaryArteryWeights",
-    "pulmonaryAirwayWeights", "sixTissueOctantBrainSegmentation",
-    "sixTissueOctantBrainSegmentationWithPriors1",
+    "pvs_shiva_t1_flair_2", "pvs_shiva_t1_flair_3", "pvs_shiva_t1_flair_4", 
+    
+    "wmh_shiva_flair_0", "wmh_shiva_flair_1", "wmh_shiva_flair_2", "wmh_shiva_flair_3",
+    "wmh_shiva_flair_4", "wmh_shiva_t1_flair_0", "wmh_shiva_t1_flair_1",
+    "wmh_shiva_t1_flair_2", "wmh_shiva_t1_flair_3", "wmh_shiva_t1_flair_4",
+    "protonLungMri", "protonLobes", "pulmonaryArteryWeights", "pulmonaryAirwayWeights",
+    "sixTissueOctantBrainSegmentation", "sixTissueOctantBrainSegmentationWithPriors1",
     "sixTissueOctantBrainSegmentationWithPriors2", "sysuMediaWmhFlairOnlyModel0",
-    "sysuMediaWmhFlairOnlyModel1", "sysuMediaWmhFlairOnlyModel2",
-    "sysuMediaWmhFlairT1Model0", 
-     "sysuMediaWmhFlairT1Model1",
+    "sysuMediaWmhFlairOnlyModel1", 
+     "sysuMediaWmhFlairOnlyModel2",
+    "sysuMediaWmhFlairT1Model0", "sysuMediaWmhFlairT1Model1",
     "sysuMediaWmhFlairT1Model2", "tidsQualityAssessment", "xrayLungOrientation",
     "xrayLungExtraction", "chexnetClassificationModel",
     "chexnetClassificationANTsXNetModel", "tb_antsxnet_model",
     "wholeHeadInpaintingFLAIR", "wholeHeadInpaintingPatchBasedT1",
     "wholeHeadInpaintingPatchBasedFLAIR", "wholeTumorSegmentationT2Flair",
     "wholeLungMaskFromVentilation", "DeepAtroposHcpT1Weights",
-    "DeepAtroposHcpT1T2Weights", "DeepAtroposHcpT1FAWeights",
+    "DeepAtroposHcpT1T2Weights", 
+     "DeepAtroposHcpT1FAWeights",
     "DeepAtroposHcpT1T2FAWeights"),
   targetFileName
 )
diff --git a/man/harvardOxfordAtlasLabeling.Rd b/man/harvardOxfordAtlasLabeling.Rd
new file mode 100644
index 0000000..6ddd3a6
--- /dev/null
+++ b/man/harvardOxfordAtlasLabeling.Rd
@@ -0,0 +1,83 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/harvardOxfordAtlasLabeling.R
+\name{harvardOxfordAtlasLabeling}
+\alias{harvardOxfordAtlasLabeling}
+\title{Subcortical and cerebellar labeling from a T1 image.}
+\usage{
+harvardOxfordAtlasLabeling(t1, doPreprocessing = TRUE, verbose = FALSE)
+}
+\arguments{
+\item{t1}{raw or preprocessed 3-D T1-weighted brain image.}
+
+\item{doPreprocessing}{perform preprocessing.  See description above.}
+
+\item{verbose}{print progress.}
+}
+\value{
+list consisting of the segmentation image and probability images for
+each label.
+}
+\description{
+Perform HOA labeling using deep learning and data from
+Morphology" number: "NIH NIMH 5R01MH112748-04". Repository:
+https://github.com/HOA-2/SubcorticalParcellations'
+}
+\details{
+The labeling is as follows:
+\itemize{
+\item{Label 1:}{Lateral Ventricle Left}
+\item{Label 2:}{Lateral Ventricle Right}
+\item{Label 3:}{CSF}
+\item{Label 4:}{Third Ventricle}
+\item{Label 5:}{Fourth Ventricle}
+\item{Label 6:}{5th Ventricle}
+\item{Label 7:}{Nucleus Accumbens Left}
+\item{Label 8:}{Nucleus Accumbens Right}
+\item{Label 9:}{Caudate Left}
+\item{Label 10:}{Caudate Right}
+\item{Label 11:}{Putamen Left}
+\item{Label 12:}{Putamen Right}
+\item{Label 13:}{Globus Pallidus Left}
+\item{Label 14:}{Globus Pallidus Right}
+\item{Label 15:}{Brainstem}
+\item{Label 16:}{Thalamus Left}
+\item{Label 17:}{Thalamus Right}
+\item{Label 18:}{Inferior Horn of the Lateral Ventricle Left}
+\item{Label 19:}{Inferior Horn of the Lateral Ventricle Right}
+\item{Label 20:}{Hippocampal Formation Left}
+\item{Label 21:}{Hippocampal Formation Right}
+\item{Label 22:}{Amygdala Left}
+\item{Label 23:}{Amygdala Right}
+\item{Label 24:}{Optic Chiasm}
+\item{Label 25:}{VDC Anterior Left}
+\item{Label 26:}{VDC Anterior Right}
+\item{Label 27:}{VDC Posterior Left}
+\item{Label 28:}{VDC Posterior Right}
+\item{Label 29:}{Cerebellar Cortex Left}
+\item{Label 30:}{Cerebellar Cortex Right}
+\item{Label 31:}{Cerebellar White Matter Left}
+\item{Label 32:}{Cerebellar White Matter Right}
+}
+
+Preprocessing on the training data consisted of:
+\itemize{
+\item n4 bias correction,
+\item brain extraction, and
+\item affine registration to HCP.
+The input T1 should undergo the same steps.  If the input T1 is the raw
+T1, these steps can be performed by the internal preprocessing, i.e. set
+\code{doPreprocessing = TRUE}
+}
+}
+\examples{
+\dontrun{
+library( ANTsRNet )
+library( keras )
+
+image <- antsImageRead( "t1.nii.gz" )
+results <- harvardOxfordAtlasLabeling( image )
+}
+}
+\author{
+Tustison NJ
+}
diff --git a/man/neuralStyleTransfer.Rd b/man/neuralStyleTransfer.Rd
index e9079a8..b2b3858 100644
--- a/man/neuralStyleTransfer.Rd
+++ b/man/neuralStyleTransfer.Rd
@@ -10,7 +10,7 @@ neuralStyleTransfer(
   initialCombinationImage = NULL,
   numberOfIterations = 10,
   learningRate = 1,
-  totalVariationWeight = 8.5e-05,
+  totalVariationWeight = 0.000085,
   contentWeight = 0.025,
   styleImageWeights = 1,
   contentLayerNames = c("block5_conv2"),
diff --git a/man/preprocessBrainImage.Rd b/man/preprocessBrainImage.Rd
index 3e9b657..b2ee63f 100644
--- a/man/preprocessBrainImage.Rd
+++ b/man/preprocessBrainImage.Rd
@@ -27,7 +27,7 @@ for intensity truncation.}
 
 \item{brainExtractionModality}{string or NULL.  Perform brain extraction
 using antsxnet tools.  One of "t1", "t1v0", "t1nobrainer", "t1combined",
-"flair", "t2", "bold", "fa", "t1infant", "t2infant", or NULL.}
+"flair", "t2", "bold", "fa", "t1infant", "t2infant", "t1threetissue", or NULL.}
 
 \item{templateTransformType}{see Details in help for \code{antsRegistration}.
 Typically "Rigid" or "Affine".}