In this study, standardization methods are used to pre-process brain MRI to generate a machine learning dataset for tumor segmentation. This method was chosen for previously documented repeatability properties as compared to more widely used normalization methods, which could potentially lead to a more generalized segmentation model. When applied to the publicly available BRATS dataset, the standardization methods performed equally as well as the normalization method used in this study, which supports further analysis of the methods beyond the highly controlled BRATS dataset.
This abstract and the presentation materials are available to members only; a login is required.