Brain segmentation based on multi-atlas and diffeomorphism guided 3D fully convolutional network ensembles

作者：

Highlights：

• We proposed a 3D CNN approach for brain structures segmentation that combines convolutional and multi-atlas features.

• An adaptive size of 3D ROI patches sampling technique to force the CNN to sufficiently learn the specific ROIs is presented.

摘要

•We proposed a 3D CNN approach for brain structures segmentation that combines convolutional and multi-atlas features.•An adaptive size of 3D ROI patches sampling technique to force the CNN to sufficiently learn the specific ROIs is presented.

论文关键词：Brain segmentation,Fully convolutional network,Multi-atlas,Diffeomorphism,Adaptive-size patches,Ensemble model

论文评审过程：Received 29 January 2020, Revised 4 October 2020, Accepted 16 February 2021, Available online 18 February 2021, Version of Record 26 February 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107904