Fully automated multi-parametric brain tumour segmentation using superpixel based classification

作者：

Highlights：

• Preprocessing is proposed to compensate inherent noise and low contrast.

• A minimal set of highly representative regional features are employed.

• The problem of class imbalance at a regional level is addressed for the first time.

• Random Forest performance is much better than Support Vector Machine.

摘要

•Preprocessing is proposed to compensate inherent noise and low contrast.•A minimal set of highly representative regional features are employed.•The problem of class imbalance at a regional level is addressed for the first time.•Random Forest performance is much better than Support Vector Machine.

论文关键词：Brain tumour,Segmentation,Localization,FLAIR,Support vector machine,Random forest classifier,BRATS

论文评审过程：Received 16 April 2018, Revised 23 September 2018, Accepted 18 October 2018, Available online 18 October 2018, Version of Record 25 October 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.10.040