Breast cancer diagnosis from histopathological images using textural features and CBIR
作者：
Highlights：
• This work proposes a method using texture features to diagnose breast cancer.
• The images are from a public database containing 400 images labeled by experts.
• Phylogenetic indexes are unprecedented form of representation of histopathological images.
• For images not labeled, we use CBIR to rank the image classes.
• The method achieved an accuracy of 95.0% in 4 classes which outperform the literature.
摘要
•This work proposes a method using texture features to diagnose breast cancer.•The images are from a public database containing 400 images labeled by experts.•Phylogenetic indexes are unprecedented form of representation of histopathological images.•For images not labeled, we use CBIR to rank the image classes.•The method achieved an accuracy of 95.0% in 4 classes which outperform the literature.
论文关键词：Breast cancer,Computer-aided diagnosis,Content-based image retrieval,Medical images,Histopathological images
论文评审过程：Received 1 August 2019, Revised 27 February 2020, Accepted 12 March 2020, Available online 22 April 2020, Version of Record 22 April 2020.
论文官网地址：https://doi.org/10.1016/j.artmed.2020.101845
