Multi-atlas segmentation and correction model with level set formulation for 3D brain MR images
作者：
Highlights：
• Bias fields and weak boundaries are difficulties for MR image segmentation.
• An accurate and efficient multi-atlas segmentation model with level set formulation for 3D human brain MR images is constructed to achieve bias correction and image segmentation simultaneously.
• 15 3D human brain MR images were segmented accurately and fast by the proposed model to obtain satisfactory segmentation and correction results for six tissues.
• Numerical segmentation results such as DICE values, RI values, GCE values, computation time and segmentation surfaces were presented.
• Sufficiently comparative experiments between the proposed model and other models have shown the efficiency and accuracy of the proposed model.
摘要
•Bias fields and weak boundaries are difficulties for MR image segmentation.•An accurate and efficient multi-atlas segmentation model with level set formulation for 3D human brain MR images is constructed to achieve bias correction and image segmentation simultaneously.•15 3D human brain MR images were segmented accurately and fast by the proposed model to obtain satisfactory segmentation and correction results for six tissues.•Numerical segmentation results such as DICE values, RI values, GCE values, computation time and segmentation surfaces were presented.•Sufficiently comparative experiments between the proposed model and other models have shown the efficiency and accuracy of the proposed model.
论文关键词：Level set formulation,Multi-atlas label fusion,Tissue segmentation,Bias correction,MR images
论文评审过程：Received 10 June 2018, Revised 16 January 2019, Accepted 25 January 2019, Available online 29 January 2019, Version of Record 16 February 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.01.031
