Main Content


Medical image segmentation using deep learning, interactive labeling app, or image processing algorithms

Image segmentation is the process of partitioning an image into regions. Semantic segmentation associates each pixel or voxel in an image with a class label that describes the meaning of an image region, such as bone, tumor, or background. You can perform medical image semantic segmentation using deep learning, the interactive Medical Image Labeler app, or image processing algorithms. Deep learning workflows require Deep Learning Toolbox™ and Computer Vision Toolbox™.


Medical Image LabelerDisplay and label 2-D and 3-D medical images


expand all

imbinarizeBinarize 2-D grayscale image or 3-D volume by thresholding
graythreshGlobal image threshold using Otsu's method
multithresh Multilevel image thresholds using Otsu’s method
otsuthreshGlobal histogram threshold using Otsu's method
adaptthreshAdaptive image threshold using local first-order statistics
grayconnectedSelect contiguous image region with similar gray values using flood-fill technique
watershedWatershed transform
activecontourSegment image into foreground and background using active contours (snakes) region growing technique
imsegfmmBinary image segmentation using fast marching method
gradientweightCalculate weights for image pixels based on image gradient
graydiffweightCalculate weights for image pixels based on grayscale intensity difference
imsegkmeansK-means clustering based image segmentation
imsegkmeans3K-means clustering based volume segmentation
superpixels2-D superpixel oversegmentation of images
superpixels33-D superpixel oversegmentation of 3-D image

Load and Prepare Training Data

groundTruthMedicalGround truth label data for medical images
changeFilePathsChange file paths in ground truth data for medical images
mergeMerge two or more groundTruthMedical objects
imageDatastoreDatastore for image data
pixelLabelDatastoreDatastore for pixel label data
randomPatchExtractionDatastoreDatastore for extracting random 2-D or 3-D random patches from images or pixel label images
combineCombine data from multiple datastores
transformTransform datastore

Import Network

importKerasLayersImport layers from Keras network
importTensorFlowLayersImport layers from TensorFlow network
importONNXLayersImport layers from ONNX network

Design Networks

fcnLayersCreate fully convolutional network layers for semantic segmentation
segnetLayersCreate SegNet layers for semantic segmentation
unet3dLayersCreate 3-D U-Net layers for semantic segmentation of volumetric images
unetLayersCreate U-Net layers for semantic segmentation
pixelClassificationLayerCreate pixel classification layer for semantic segmentation
dicePixelClassificationLayerCreate pixel classification layer using generalized Dice loss for semantic segmentation

Segment Images

semanticsegSemantic image segmentation using deep learning
labeloverlayOverlay label matrix regions on 2-D image
volshowDisplay volume
jaccardJaccard similarity coefficient for image segmentation
diceSørensen-Dice similarity coefficient for image segmentation
bfscoreContour matching score for image segmentation


Related Information