A novel learning-based dissimilarity metric for rigid and non-rigid medical image registration by using Bhattacharyya Distances
作者：
Highlights：
• We propose a novel learning-based dissimilarity metric for both rigid and non-rigid medical image registrations.
• The novel dissimilarity metric is based on the Bhattacharyya Distance.
• A lot of experiments were done to evaluate our method in term of accuracy and robustness.
• We compared the proposed method with state-of-the-art methods of both rigid and non-rigid image registration.
• Our method is more robust against different challenging cases with consistently higher accuracy.
摘要
Highlights•We propose a novel learning-based dissimilarity metric for both rigid and non-rigid medical image registrations.•The novel dissimilarity metric is based on the Bhattacharyya Distance.•A lot of experiments were done to evaluate our method in term of accuracy and robustness.•We compared the proposed method with state-of-the-art methods of both rigid and non-rigid image registration.•Our method is more robust against different challenging cases with consistently higher accuracy.
论文关键词：Bhattacharyya Distance,Rigid and non-rigid medical image registration,Learning-based dissimilarity metric
论文评审过程：Received 17 March 2016, Revised 7 September 2016, Accepted 8 September 2016, Available online 9 September 2016, Version of Record 17 September 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.09.004
