Region Growing and Level Set Compound for Hippocampus Segmentation

In any automatic detection of Alzheimer's disease using structural Magnetic Resonance Images, accurate segmentation of hippocampus is significant. A new method is proposed to segment the hippocampus from the Magnetic Resonance Image by combining the region growing and level set methods. The first step is to segment the hippocampus using region growing from the seed point. Second step is to fine tune the segmented hippocampus using level set. This overcomes the draw backs of potential leakage in region growing method for hippocampus segmentation. Quantitative analysis shows a close correlation between the proposed segmentation and the human expert segmentation.