Multiple instance learning for histopathological breast cancer image classification

作者：

Highlights：

• Multiple Instance Learning (MIL) is a framework to deal with weakly supervised data.

• We investigate the relevance of MIL for histopathological cancer image classification.

• Benchmark of MIL methods on a public dataset shows the interest of non parametric MIL.

• Compared to single instance learning, MIL can bring some insight on such application.

摘要

•Multiple Instance Learning (MIL) is a framework to deal with weakly supervised data.•We investigate the relevance of MIL for histopathological cancer image classification.•Benchmark of MIL methods on a public dataset shows the interest of non parametric MIL.•Compared to single instance learning, MIL can bring some insight on such application.

论文关键词：Biomedical image processing,Breast cancer,Histopathology,Image classification,Multiple Instance Learning

论文评审过程：Received 5 June 2018, Revised 31 August 2018, Accepted 23 September 2018, Available online 24 September 2018, Version of Record 27 September 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.09.049