Reference

Classification/regression models
createAlexNetModel2D()	2-D implementation of the AlexNet deep learning architecture.
createAlexNetModel3D()	3-D implementation of the AlexNet deep learning architecture.
createDenseNetModel2D()	2-D implementation of the DenseNet deep learning architecture.
createDenseNetModel3D()	3-D implementation of the DenseNet deep learning architecture.
createGoogLeNetModel2D()	2-D implementation of the GoogLeNet deep learning architecture.
createResNetModel2D()	2-D implementation of the ResNet deep learning architecture.
createResNetModel3D()	3-D implementation of the ResNet deep learning architecture.
createWideResNetModel2D()	2-D implementation of the Wide ResNet deep learning architecture.
createWideResNetModel3D()	3-D implementation of the Wide ResNet deep learning architecture.
createVggModel2D()	2-D implementation of the VGG deep learning architecture.
createVggModel3D()	3-D implementation of the VGG deep learning architecture.
createResNetWithSpatialTransformerNetworkModel2D()	2-D implementation of the ResNet deep learning architecture with a preceding spatial transformer network layer.
createResNetWithSpatialTransformerNetworkModel3D()	3-D implementation of the ResNet deep learning architecture with a preceding spatial transformer network layer.
createSimpleClassificationWithSpatialTransformerNetworkModel2D()	2-D implementation of the spatial transformer network.
createSimpleClassificationWithSpatialTransformerNetworkModel3D()	3-D implementation of the spatial transformer network.
createSimpleFullyConvolutionalNeuralNetworkModel3D()	Implementation of the "SCFN" architecture for Brain/Gender prediction
Object localization models
createSsd7Model2D()	2-D implementation of the SSD 7 deep learning architecture.
createSsd7Model3D()	3-D implementation of the SSD 7 deep learning architecture.
createSsdModel2D()	2-D implementation of the SSD deep learning architecture.
createSsdModel3D()	3-D implementation of the SSD deep learning architecture.
SSD-specific helper functions
AnchorBoxLayer2D	Anchor box layer for SSD architecture (2-D).
AnchorBoxLayer3D	Anchor box layer for SSD architecture (3-D).
L2NormalizationLayer2D	L2 2-D normalization layer for SSD300/512 architecture.
L2NormalizationLayer3D	L2 3-D normalization layer for SSD300/512 architecture.
LossSSD	Loss function for the SSD deep learning architecture.
convertCoordinates()	Convert coordinates to/from min/max representation from/to centroids/width
decodeSsd2D()	Decoding function for 2-D Y_train
decodeSsd3D()	Decoding function for 3-D Y_train
drawRectangles()	Plotting function for 2-D object detection visualization.
encodeSsd2D()	Encoding function for 2-D Y_train
encodeSsd3D()	Encoding function for 3-D Y_train
jaccardSimilarity()	Jaccard similarity between two sets of boxes.
Segmentation
createUnetModel2D()	2-D implementation of the U-net deep learning architecture.
createUnetModel3D()	3-D image segmentation implementation of the U-net deep learning architecture.
createResUnetModel2D()	2-D implementation of the Resnet + U-net deep learning architecture.
createResUnetModel3D()	3-D implementation of the Resnet + U-net deep learning architecture.
createDenseUnetModel2D()	2-D implementation of the dense U-net deep learning architecture.
createDenseUnetModel3D()	3-D implementation of the dense U-net deep learning architecture.
createNoBrainerUnetModel3D()	Implementation of the "NoBrainer" U-net architecture
createHippMapp3rUnetModel3D()	Implementation of the "HippMapp3r" U-net architecture
createSysuMediaUnetModel2D()	Implementation of the sysu_media U-net architecture
createHypothalamusUnetModel3D()	3-D u-net architecture for hypothalamus segmentation
Segmentation-specific helper functions
decodeUnet()	Decoding function for the u-net prediction outcome
encodeUnet()	One-hot encoding function
ScaleLayer	Custom scale layer
Super-resolution
createDeepDenoiseSuperResolutionModel2D()	2-D implementation of the deep denoise image super resolution architecture.
createDeepDenoiseSuperResolutionModel3D()	3-D implementation of the deep denoise image super resolution architecture.
createDenoisingAutoEncoderSuperResolutionModel2D()	2-D implementation of the denoising autoencoder image super resolution architecture.
createDenoisingAutoEncoderSuperResolutionModel3D()	3-D implementation of the denoising autoencoder image super resolution architecture.
createExpandedSuperResolutionModel2D()	2-D implementation of the expanded image super resolution architecture.
createExpandedSuperResolutionModel3D()	3-D implementation of the expanded image super resolution architecture.
createImageSuperResolutionModel2D()	2-D implementation of the image super resolution deep learning architecture.
createImageSuperResolutionModel3D()	3-D implementation of the image super resolution deep learning architecture.
createResNetSuperResolutionModel2D()	2-D implementation of the ResNet image super resolution architecture.
createResNetSuperResolutionModel3D()	3-D implementation of the ResNet image super resolution architecture.
createDeepBackProjectionNetworkModel2D()	2-D implementation of the deep back-projection network.
createDeepBackProjectionNetworkModel3D()	3-D implementation of the deep back-projection network.
applySuperResolutionModelToImage()	Apply a pretrained model for super resolution.
Super-resolution helper functions
MSE()	Mean square error of a single image or between two images.
MAE()	Mean absolute error of a single image or between two images.
PSNR()	Peak signal-to-noise ratio between two images.
SSIM()	Structural similarity index (SSI) between two images.
Registration and transforms
SpatialTransformerLayer2D	Spatial transformer layer (2-D)
SpatialTransformerLayer3D	Spatial transfomer layer (3-D)
Generative adverserial networks (GAN)
VanillaGanModel	Vanilla GAN model
DeepConvolutionalGanModel	Deep convolutional GAN (DCGAN) model
WassersteinGanModel	Wasserstein GAN model
ImprovedWassersteinGanModel	Improved Wasserstein GAN model
CycleGanModel	Cycle GAN model
SuperResolutionGanModel	Super resolution GAN model
InstanceNormalizationLayer	Creates an instance normalization layer
layer_instance_normalization()	Instance normalization layer
Attention
AttentionLayer2D	Attention layer (2-D)
AttentionLayer3D	Attention layer (3-D)
layer_attention_2d()	Attention layer (2-D)
layer_attention_3d()	Attention layer (3-D)
EfficientAttentionLayer2D	Efficient attention layer (2-D)
EfficientAttentionLayer3D	Efficient attention layer (3-D)
layer_efficient_attention_2d()	Efficient attention layer (2-D)
layer_efficient_attention_3d()	Efficient attention layer (3-D)
Deep embedded clustering (DEC)
DeepEmbeddedClusteringModel	Deep embedded clustering (DEC) model class
ClusteringLayer	Clustering layer for Deep Embedded Clustering
createAutoencoderModel()	Function for creating a symmetric autoencoder model.
createConvolutionalAutoencoderModel2D()	Function for creating a 2-D symmetric convolutional autoencoder model.
createConvolutionalAutoencoderModel3D()	Function for creating a 3-D symmetric convolutional autoencoder model.
Mixture density networks
MixtureDensityNetworkLayer	Mixture density network layer
getMixtureDensityLossFunction()	Returns a loss function for the mixture density.
getMixtureDensitySamplingFunction()	Returns a sampling function for the mixture density.
getMixtureDensityMseAccuracyFunction()	Returns a MSE accuracy function for the mixture density.
splitMixtureParameters()	Splits the mixture parameters.
sampleFromCategoricalDistribution()	Sample from a categorical distribution
sampleFromOutput()	Sample from a distribution
mixture_density_network_softmax()	Softmax function for mixture density with temperature adjustment
Custom metrics/losses
binary_dice_coefficient()	Dice function for binary segmentation problems
multilabel_dice_coefficient()	Dice function for multilabel segmentation problems
peak_signal_to_noise_ratio()	Function to calculate peak-signal-to-noise ratio.
pearson_correlation_coefficient()	Function for Pearson correlation coefficient.
categorical_focal_gain()	Function for categorical focal gain
categorical_focal_loss()	Function for categorical focal loss
weighted_categorical_crossentropy()	Function for weighted categorical cross entropy
maximum_mean_discrepancy()	Function for maximum-mean discrepancy
Custom activations
layer_activation_log_softmax()	Log softmax layer
peak_signal_to_noise_ratio()	Function to calculate peak-signal-to-noise ratio.
pearson_correlation_coefficient()	Function for Pearson correlation coefficient.
categorical_focal_gain()	Function for categorical focal gain
categorical_focal_loss()	Function for categorical focal loss
Data augmentation
basisWarp()	Generate a deformable map from basis
randomImageTransformAugmentation()	Apply random transforms to a predictor / outcome training image set
randomImageTransformBatchGenerator	Random image transformation batch generator
randomImageTransformParametersAugmentation()	Generate transform parameters and transformed images
randomImageTransformParametersBatchGenerator	Random image transform parameters batch generator
randomlyTransformImageData()	Randomly transform image data (optional: with corresponding segmentations).
dataAugmentation()	Randomly transform image data.
histogramWarpImageIntensities()	Transform image intensities based on histogram mapping.
simulateBiasField()	Simulate random bias field
Misc
preprocessBrainImage()	Basic preprocessing pipeline for T1-weighted brain MRI
extractImagePatches()	Extract 2-D or 3-D image patches.
cropImageCenter()	Crop the center of an image.
padOrCropImageToSize()	Pad or crop image to a specified size
padImageByFactor()	Pad an image based on a factor.
reconstructImageFromPatches()	Reconstruct image from a list of patches.
getPretrainedNetwork()	getPretrainedNetwork
getANTsXNetData()	getANTsXNetData
regressionMatchImage()	Image intensity normalization using linear regression.
ResampleTensorLayer2D	Creates a resample tensor layer (2-D)
ResampleTensorLayer3D	Creates a resampled tensor (to fixed size) layer (3-D)
layer_resample_tensor_2d()	Creates a resampled tensor (to fixed size) layer (2-D)
layer_resample_tensor_3d()	Resampling a spatial tensor (3-D).
ResampleTensorToTargetTensorLayer2D	Creates a resampled tensor (to target tensor) layer (2-D)
ResampleTensorToTargetTensorLayer3D	Creates a resampled tensor (to target tensor) layer (3-D)
Applications
brainExtraction()	Brain extraction
deepAtropos()	Six tissue segmentation
corticalThickness()	Cortical thickness using deep learning
longitudinalCorticalThickness()	Longitudinal cortical thickness using deep learning
lungExtraction()	Lung extraction
brainAge()	BrainAGE
hippMapp3rSegmentation()	hippMapp3rSegmentation
sysuMediaWmhSegmentation()	White matter hyperintensity segmentation
claustrumSegmentation()	Claustrum segmentation
hypothalamusSegmentation()	Hypothalamus segmentation
deepFlash()	Hippocampal/Enthorhinal segmentation using "Deep Flash"
deepFlashDeprecated()	Hippocampal/Enthorhinal segmentation using "Deep Flash"
desikanKillianyTourvilleLabeling()	Cortical and deep gray matter labeling using Desikan-Killiany-Tourville
mriSuperResolution()	Super-resolution for MRI
neuralStyleTransfer()	Neural transfer style
tidNeuralImageAssessment()	Perform MOS-based assessment of an image.
elBicho()	Functional lung segmentation.
arterialLesionSegmentation()	Arterial lesion segmentation