The final output of SXT is a 3D structural model of the specimen. This model has to be segmented into its components, such as the nucleus, mitochondria, and a host of other subcellular features. This has long been a time-consuming process. We are now automating this process by applying Machine Learning.