TomoSAM: a 3D Slicer extension using SAM for tomography segmentation

For this reason, many deep learning techniques have been proposed to perform the segmentation task, but they usually require a large amount of labeled data to train, as well as considerable computational resources. Our work is motivated by the objective of modeling Thermal Protection Systems (TPS), in order to estimate their material properties and predict their response to the harsh environmental conditions experienced during atmospheric entry. The Porous Microstructure Analysis (PuMA) software [1] was developed to provide a robust and efficient framework for the digital characterization of 3D microstructures.