Weighted Shape-Based Averaging With Neighborhood Prior Model for Multiple Atlas Fusion-Based Medical Image Segmentation

@article{Gorthi2013WeightedSA,
  title={Weighted Shape-Based Averaging With Neighborhood Prior Model for Multiple Atlas Fusion-Based Medical Image Segmentation},
  author={Sai Subrahmanyam Gorthi and Meritxell Bach Cuadra and Pierre-Alain Tercier and Abdelkarim Allal and Jean-Philippe Thiran},
  journal={IEEE Signal Processing Letters},
  year={2013},
  volume={20},
  pages={1034-1037}
}
In medical imaging, merging automated segmentations obtained from multiple atlases has become a standard practice for improving the accuracy. In this letter, we propose two new fusion methods: “Global Weighted Shape-Based Averaging” (GWSBA) and “Local Weighted Shape-Based Averaging” (LWSBA). These methods extend the well known Shape-Based Averaging (SBA) by additionally incorporating the similarity information between the reference (i.e., atlas) images and the target image to be segmented. We… CONTINUE READING