Edge U-Net: Brain tumor segmentation using MRI based on deep U-Net model with boundary information

作者：

Highlights：

• CLAHE is used to enhance the MRI contrast.

• Edge-deep model developed to achieve high accurate segmentation.

• Fusing the MRI boundaries features and MRI features by EGB module.

• Combination of the cross-entropy and boundaries loss functions are used.

• Therefore, Dice score of our framework is higher than other state-of-the-art models.

摘要

•CLAHE is used to enhance the MRI contrast.•Edge-deep model developed to achieve high accurate segmentation.•Fusing the MRI boundaries features and MRI features by EGB module.•Combination of the cross-entropy and boundaries loss functions are used.•Therefore, Dice score of our framework is higher than other state-of-the-art models.

论文关键词：Brain tumor segmentation,Boundary information,Convolutional neural network,MRI,Deep learning,Contrast limited adaptive histogram equalisation

论文评审过程：Received 7 July 2022, Revised 8 September 2022, Accepted 11 September 2022, Available online 22 September 2022, Version of Record 28 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118833