Meeting Banner
Abstract #1274


Alexey Shukelovich 1,2 , Pierre-Yves Baudin 1 , Noura Azzabou 1 , Pierre G Carlier 1 , Jean-Marc Boisserie 1 , and Julien LeLouer 1

1 AIM-CEA Institut de Myologie, Laboratoire RMN, Paris, France, 2 The United Institute of Informatics Problems of the National Academy of Sciences of Belarus, Minsk, Belarus

In this work we present a user-friendly interactive segmentation tool that is handy and suitable for clinicians and is developed to improve, accelerate and facilitate MRI muscle segmentation process. Segmentation software is built on the top of ITK-SNAP and incorporates a new segmentation tool based on a robust semi-automatic random walker algorithm. Experimental segmentation was done on a human thigh muscle MRI dataset and compared to a manual segmentation using relative volume differences and Dice coefficients. We have achieved a sufficient acceleration in segmentation process with minor loss of segmentation quality.

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

Join Here