Automated Sample Points Algorithm

Technology #2008-026

An algorithm has been developed to automatically define intensity properties for key tissues in magnetic resonance brain images. The algorithm is completely automated in that it does not require that the user provide samples of any tissues, and it does not require that the user provide a brain parenchyma mask. The algorithm is additionally robust in that it is applicable to images acquired using a range of acquisition protocols, and which have been acquired for a range of disease processes. The algorithm should serve as the foundation for examples of Al-supervised image processing.