Anette Karlsson1, 2, Olof Dahlqvist Leinhard, 23, Anna Vallin2, Thobias Romu1, 2, Magnus Borga1, 2
1Department of Biomedical Engineering, Linkping University, Linkping, Sweden; 2Center for Medical Image Science and Visualization (CMIV), Linkping University, Linkping, Sweden; 3Depts of Radiation Physics, Linkping University and Radiation Physics, UHL County Council of Ostergotland, Linkping, Sweden
We have previously developed methods where intensity normalized water and fat separated images can be used for fat quantification. This approach has now been developed for performing automatic muscle quantification in 3D whole body volumes. A water volume of a subject is used as a prototype where the muscles are segmented manually. The prototype volume is deformed to match the water volume of the target image using non-rigid registration. The muscle mask of the prototype is also deformed giving an automatic muscle mask to the target image which can be used for quantification.