Meeting Banner
Abstract #0753

Atlas-Free Brain Tissue Segmentation Using a Single T1-Weighted MRI Acquisition

Tobias Kober1, Alexis Roche1, 2, Oscar Esteban3, 4, Subrahmanyam Gorthi4, Delphine Ribes5, Meritxell Bach-Cuadra2, 4, Reto Meuli2, Gunnar Krueger1, 2

1Advanced Clinical Imaging Technology, Siemens Healthcare Sector IM&WS S, Lausanne, Switzerland; 2Centre Hospitalier Universitaire Vaudois and University of Lausanne, Switzerland; 3Biomedical Image Technology (BIT), Universidad Politcnica de Madrid, Madrid, Spain; 4Signal Processing Laboratory (LTS5), cole Polytechnique Fdrale de Lausanne, Lausanne, Switzerland; 5ARTORG Center for Computer Aided Surgery, University of Bern, Switzerland

Typical brain tissue segmentation algorithms use prior knowledge in the form of pre-segmented templates, i.e. atlases. This can lead to erroneous segmentation results if the actual image differs too much from the template, e.g. due to a too big age difference or a disease. In this work, the homogenous T1-contrast of the MP2RAGE sequence is combined with a Dixon water/fat imaging approach. It is shown that a single MP2RAGE-Dixon acquisition provides enough information for reproducible atlas-free brain tissue segmentation.

Abstract PDF Presentation Video

Keywords

abstract accurate achieve achieved acquisition addition additional aging aided alterations anal anatomically assessment atlas automatic axial background bandwidth biomedical brain brains calculating changing channel classes classified combination combine combined combining confirming conjunction connected considerably constraints contrast coronal crucial decomposition degradation deviate diseased double drawing echoes employing enforced enough except exemplary exploit extraction final fluid foundations free gained green hard head healthy heavily hole homogeneous illustrate implemented imposed improves included indicate input interface introduced inversion laboratory largest limitations links little loss mater material mathematical model morphology mostly muscle noise notably noted operations pattern potential preliminary prior processing properties proposed protocol providing pure race radiology rage rages random readout readouts recovered reduced reliable rely removal rendering reproducible rescans robust root scanned sector segmentation segmentations segmented separation seven since skull slices step stripping studies subject subjects suffers supported supports surgery technology template tissue tissues took topological trio typically uniform view volume volumes volunteers yielding