On Nonparametric Markov Random Field Estimation for Fast Automatic Segmentation of MRI Knee Data
Filip Kor, David C. Schneider, Wolfgang Förstner · 2010
We present a fast automatic reproducible method for 3d se- mantic segmentation of magnetic resonance images of the knee. We for- mulate a single global model that allows to jointly segment all classes. The model estimation was performed automatically without manual in- teraction and parameter tuning. The segmentation of a magnetic reso- nance image with 11 Mio voxels took approximately one minute. Our labeling results by far do not reach the performance of complex state of the art approaches designed to produce clinically relevant results. Our results could potentially be useful for rough visualization or initializa- tion of computationally demanding methods. Our main contribution is to provide insights in possible strategies when employing global statisti- cal models.