Segmentation Using Neural Nets

U-Net is a convolutional neural network architecture for semantic image segmentation. This particular network has been trained on data provided by the ISBI cell tracking challenge 2015. The U-Net is well suited to segment glioblastoma-astrocytoma cells in microscopy images.

Import the neural network from the Wolfram Neural Net Repository.

The following function utilizes U-Net to obtain cell segmentations in images of arbitrary size.

Create a cell segmentation mask.

Highlight the resulting cell segmentation.