Computational analysis of histological images from hematoxylin and eosin-stained oral epithelial dysplasia tissue sections

作者：

Highlights：

• A new approach for OED cell nuclei detection for the grades of the lesion.

• Association of morphological and texture features with a polynomial classifier.

• A dataset of images from mice tongues was employed in this study.

• AUC values up to 0.96 were obtained for classifying the lesion grades.

• A novel method for automatic grading of oral dysplasia.

摘要

•A new approach for OED cell nuclei detection for the grades of the lesion.•Association of morphological and texture features with a polynomial classifier.•A dataset of images from mice tongues was employed in this study.•AUC values up to 0.96 were obtained for classifying the lesion grades.•A novel method for automatic grading of oral dysplasia.

论文关键词：Dysplasia,Convolutional neural network,Oral cavity,Polynomial classifier,Histological image

论文评审过程：Received 2 March 2021, Revised 14 December 2021, Accepted 23 December 2021, Available online 12 January 2022, Version of Record 19 January 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116456