Intensity population based unsupervised hemorrhage segmentation from brain CT images

作者：

Highlights：

• Complete automatic segmentation process.

• Skull and background removal to enhance accuracy and make the system fast.

• Automatic selection of threshold for hemorrhage from intensity population graph.

• Accuracy enhancement of segmentation through image restoration.

• Result is compared with established and known methods.

摘要

•Complete automatic segmentation process.•Skull and background removal to enhance accuracy and make the system fast.•Automatic selection of threshold for hemorrhage from intensity population graph.•Accuracy enhancement of segmentation through image restoration.•Result is compared with established and known methods.

论文关键词：Hemorrhage,CT,Segmentation,CAD,Thresholding,Brain

论文评审过程：Received 12 June 2017, Revised 16 December 2017, Accepted 17 December 2017, Available online 21 December 2017, Version of Record 30 December 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.12.032