Abstract #5077

A T2 Template Map from a Healthy Cohort to Identify Localized Anomalies in Single Subjects

Tom Hilbert^1,2,3, Alexis Roche^1,2,3, Cristina Granziera^4,5, Guillaume Bonnier⁵, Kieran O’Brien^6,7, Tony Stöcker⁸, Pavel Falkovskiy^1,2,3, Reto Meuli², Jean-Philippe Thiran^2,3, Gunnar Krueger^2,3,9, and Tobias Kober^1,2,3

¹Advanced Clinical Imaging Technology, Siemens Healthcare AG, Lausanne, Switzerland, ²Department of Radiology, University Hospital (CHUV), Lausanne, Switzerland, ³LTS5, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ⁴Neuroimmunology Unit, Neurology, Department of Clinical Neurosciences, Centre Hospitalier Universitaire Vaudois (CHUV) and University of Lausanne (UNIL), Lausanne, Switzerland, ⁵Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States, ⁶Center for Advanced Imaging, the University of Queensland, Brisbane, Australia, ⁷Siemens Healthcare Pty Ltd, Brisbane, Australia, ⁸German Center for Neurodegenerative Diseases, Bonn, Germany, ⁹Siemens Medical Solutions USA, Inc., Boston, MA, United States

We construct a database of normal T2 values by spatially normalizing quantitative maps from healthy subjects into a common space. A low standard deviation across all subjects demonstrates good reproducibility of the T2 values in white matter and deep grey matter. Additionally we adopt a standard voxel-based procedure that compares the quantitative T2 map of a patient to the database and test it on three multiple sclerosis datasets. The obtained z-score maps show that white matter lesions can be detected in the limits of the available resolution and the applied smoothing.

This abstract and the presentation materials are available to members only; a login is required.

Join Here